Professional Documents
Culture Documents
Dynamics and Control of A Battery Inverter Single-Phase Induction Generator System
Dynamics and Control of A Battery Inverter Single-Phase Induction Generator System
Obasohan I. Omozusi
The steady-state model for the system was developed for impedance and motor
loads using the harmonic balance technique. Experimental results for different impedance
and motor loads compare favorably with the steady-state model calculation.
VITA
College, Benin-City, Nigeria, and graduated in June 1986. He received his B.Sc. in
_______________________
A Thesis
Presented to
by
Obasohan I. Omozusi
________________________
In Partial Fulfillment
MASTER OF SCIENCE
Electrical Engineering
___________________________
December 1998
by
Obasohan I. Omozusi
Graduate Advisory Committee:
________________________ _______
Chairperson date
________________________ ________
Member date
________________________ ________
Member date
Approved for the Faculty:
Date
177
In presenting this thesis in partial fulfillment of the requirements for a Master of Science degree
at Tennessee Technological University, I agree that the University Library shall make it available to
borrowers under rules of the Library. Brief quotations from this thesis are allowable without special
Permission for extensive quotation from or reproduction of thesis may be granted by my major
professor when the proposed use of the material is for scholarly purposes. Any copying or use of the
material in this thesis for financial gain shall not be allowed without my written permission.
Signature ________________________________
Date ____________________________________
178
DEDICATION
ACKNOWLEDGMENTS
I would like to express my appreciation to Dr. Olorunfemi Ojo, my major professor and chairman
of the advisory committee, for his valuable direction and assistance. I would also like to thank Dr.
Charles Hickman a member of my advisory committee for his assistance during this work. I also want
I express my gratitude to Dr. Esther Ososanya and Dr. Ghadir Radman for their assistance. I am
grateful to Dr. P. K. Rajan, Dr. Ken Purdy, Dr. M. O. Omoigui, Dr. Jeffrey R. Austen, Prof. Eusebisu
J. Doedel, Antonio Ginart, Zhiqing Wu, Sandy Garrison, Helen Knott, Keith Jones, Conard F.
Murray, L. V. Randolph, Patricia A. Roberts, and Joel Seber for their assistance. I acknowledge the
financial support of the Center for Electric Power at Tennessee Technological University for this
research.
180
TABLE OF CONTENTS
....................................................................................................................................................................
............................................................................................................................................................Page
..................................................................................................................................................................x
LIST OF FIGURES....................................................................................................................................
................................................................................................................................................................ xi
Chapter
1. Introduction ............................................................................................................................................
..................................................................................................................................................................1
1.1. Introduction.................................................................................................................................
.........................................................................................................................................................1
.........................................................................................................................................................2
.........................................................................................................................................................7
2.1. Introduction 12
.......................................................................................................................................................13
.......................................................................................................................................................13
.......................................................................................................................................................16
.......................................................................................................................................................16
181
.......................................................................................................................................................24
.......................................................................................................................................................26
................................................................................................................................................................39
................................................................................................................................................................39
.......................................................................................................................................................43
Chapter Page
.......................................................................................................................................................48
................................................................................................................................................................50
.......................................................................................................................................................51
.......................................................................................................................................................52
................................................................................................................................................................54
................................................................................................................................................................59
................................................................................................................................................................70
................................................................................................................................................................71
................................................................................................................................................................82
Chapter Page
4.5 Parametric Studies for the Battery Inverter Single-Phase Induction Generator
with Impedance Load ...................................................................................................................
..............................................................................................................................................................106
4.6. Transient and Dynamic Performance for the System Feeding an Impedance
Load 113
..............................................................................................................................................................126
..............................................................................................................................................................127
5.3.1. Battery Inverter SPIG System Feeding SPIM Load (Linear Region) 132
..............................................................................................................................................................140
5.5 Parametric Studies for the Battery Inverter Single-Phase Induction Generator
with SPIM Load ...........................................................................................................................
..............................................................................................................................................................147
5.6. Transient and Dynamic Performance for the System feeding a SPIM Load 149
Chapter Page
..............................................................................................................................................................173
..............................................................................................................................................................174
..............................................................................................................................................................177
..............................................................................................................................................................181
..............................................................................................................................................................185
7.1. Conclusions.................................................................................................................................
.....................................................................................................................................................185
.....................................................................................................................................................186
REFERENCES ...........................................................................................................................................
..............................................................................................................................................................187
APPENDIX ................................................................................................................................................
..............................................................................................................................................................191
VITA ..........................................................................................................................................................
..............................................................................................................................................................194
185
LIST OF TABLES
Table Page
2.1. Switching state of the unipolar PWM and the corresponding voltage levels ..............................
.......................................................................................................................................................26
2.2. Switching state of the modified bipolar PWM and the corresponding voltage
levels ...........................................................................................................................................
.......................................................................................................................................................28
.......................................................................................................................................................60
.......................................................................................................................................................61
186
LIST OF FIGURES
Figure Page
.........................................................................................................................................................8
.......................................................................................................................................................13
.......................................................................................................................................................14
.......................................................................................................................................................15
.......................................................................................................................................................16
.......................................................................................................................................................17
.......................................................................................................................................................25
.......................................................................................................................................................27
................................................................................................................................................................36
.......................................................................................................................................................36
.......................................................................................................................................................37
.......................................................................................................................................................37
................................................................................................................................................................40
.......................................................................................................................................................57
3.3. Equivalent circuit of single-phase induction machine with core loss .........................................
.......................................................................................................................................................57
Figure Page
3.4. Equivalent circuit of single-phase induction machine with Ids = 0, and ωr = 0 ...........................
.......................................................................................................................................................61
3.5. Equivalent circuit of single-phase induction machine with Iqs = 0, and ωr = 0 ...........................
.......................................................................................................................................................63
3.6. Equivalent circuit of single-phase induction machine with Ids = 0, and ωr = ωe..........................
.......................................................................................................................................................64
3.7. Equivalent circuit of single-phase induction machine with Iqs = 0, and ωr = ωe..........................
.......................................................................................................................................................66
3.8. Experimental values q-d magnetizing inductances vs peak input voltage ..................................
.......................................................................................................................................................68
3.9. Experimental values of q-d core loss resistance vs peak input voltage.......................................
.......................................................................................................................................................69
3.10. Experimental values sum of q-d axis leakage inductances vs peak input
188
voltage..........................................................................................................................................69
.......................................................................................................................................................70
4.3. The q-d equivalent circuit of the battery-PWM inverter generator system .................................
.......................................................................................................................................................72
Figure Page
4.40. Contour plot of main winding current, [A] as a variation of generator rotor
speed (per unit) and load capacitor Cq. .....................................................................................
.....................................................................................................................................................107
Figure Page
191
4.43. Contour plot of auxiliary winding voltage, [V] as a variation of generator rotor
speed (per unit) and load capacitor Cq. .....................................................................................
.....................................................................................................................................................109
4.44. Contour plot of auxiliary winding current, [A] as a variation of generator rotor
speed (per unit) and load capacitor Cq. .....................................................................................
.....................................................................................................................................................110
4.45. Contour plot of efficiency as a variation of generator rotor speed (per unit)
and load capacitor Cq. ...............................................................................................................
.....................................................................................................................................................110
4.48. Contour plot of efficiency as a variation of generator rotor speed (per unit)
and load capacitor Cq. ...............................................................................................................
.....................................................................................................................................................112
4.51. Main winding current start-up waveform. Modulation index = 0.8, load
impedance = 20 Ohms, rotor speed = 1860 rpm. .....................................................................
.....................................................................................................................................................115
4.55. Changes in the values of the load impedance. Modulation index = 0.8,
and rotor speed = 1840 rpm. ....................................................................................................
.....................................................................................................................................................118
Figure Page
4.61. Changes in the values of generator rotor speeds. Modulation index = 0.8,
and load resistance = 40 Ohms. ...............................................................................................
.....................................................................................................................................................122
4.63. Main winding current waveform response to changes in generator rotor speed.
Modulation index = 0.8, and load resistance = 40 Ohms. ........................................................
.....................................................................................................................................................123
.....................................................................................................................................................126
5.2. Schematic diagram of the system with induction motor load .....................................................
.....................................................................................................................................................128
193
5.3. The q-d equivalent circuit of the battery-PWM inverter generator system with
single-phase induction motor load ..............................................................................................
.....................................................................................................................................................129
Figure Page
5.18. Measured and calculated motor input voltage as a function of motor speed.
Constant generator rotor speed = 1840 rpm ..............................................................................
.....................................................................................................................................................143
5.23. Predicted motor torque as a function of motor rotor speed. Constant generator
rotor speed = 1840 rpm .............................................................................................................
.....................................................................................................................................................146
5.24. Contour plot of torque, [Nm] as a variation of generator rotor speed (per unit)
and motor rotor speed (per unit)................................................................................................
.....................................................................................................................................................150
5.25. Contour plot of motor voltage, [V] as a variation of generator rotor speed
(per unit) and motor rotor speed (per unit)................................................................................
.....................................................................................................................................................150
5.26. Contour plot of output power, [W] as a variation of generator rotor speed
(per unit) and motor rotor speed (per unit)................................................................................
.....................................................................................................................................................151
5.27. Contour plot of battery current, [A] as a variation of generator rotor speed
(per unit) and motor rotor speed (per unit)................................................................................
.....................................................................................................................................................151
5.36. Changes in the values of the load torque. Modulation index = 2.0
and rotor generator speed = 1840 rpm ......................................................................................
.....................................................................................................................................................157
5.40. Generator main winding current waveform response to changes in load torque.
Modulation index = 2.0 and rotor generator speed = 1840 rpm................................................
.....................................................................................................................................................159
5.44. Changes in the values of generator rotor speeds. Modulation index = 2.0
and Load Torque = 0.25 Nm.....................................................................................................
.....................................................................................................................................................162
5.53. Changes in the values of the load torque. Modulation index = 1.2 and
rotor generator speed = 1840 rpm .............................................................................................
.....................................................................................................................................................168
5.54. Motor input voltage waveform. Modulation index = 1.2 and rotor generator
speed = 1840 rpm......................................................................................................................
.....................................................................................................................................................168
5.55. Motor main winding current waveform. Modulation index = 1.2 and
rotor generator speed = 1840 rpm .............................................................................................
.....................................................................................................................................................169
5.56. Motor speed waveform. Modulation index = 1.2 and rotor generator
speed = 1840 rpm......................................................................................................................
.....................................................................................................................................................169
5.57. Motor torque waveform. Modulation index = 1.2 and rotor generator
197
.....................................................................................................................................................170
5.58. Generator main winding current waveform. Modulation index = 1.2 and
rotor generator speed = 1840 rpm .............................................................................................
.....................................................................................................................................................170
5.59. Generator auxiliary winding current waveform. Modulation index = 1.2 and
rotor generator speed = 1840 rpm .............................................................................................
.....................................................................................................................................................171
5.60. Battery current waveform. Modulation index = 1.2 and rotor generator
speed = 1840 rpm......................................................................................................................
.....................................................................................................................................................172
5.61. Generator torque waveform. Modulation index = 1.2 and rotor generator
speed = 1840 rpm......................................................................................................................
.....................................................................................................................................................172
.....................................................................................................................................................173
.....................................................................................................................................................174
6.3. The q-d equivalent circuit of the capacitor-PWM inverter generator system
with impedance load ..................................................................................................................
.....................................................................................................................................................175
Figure Page
6.4. Self-excitation for the system without battery. Calculated load resistance vs
generator rotor speed..................................................................................................................
.....................................................................................................................................................182
6.5. Self-excitation for the system without battery. Contour plot of load resistance,
[Ω] as a variation of generator rotor speed and load capacitor ...................................................
.....................................................................................................................................................183
6.6. Self-excitation for the system without battery. Contour plot of load resistance,
[Ω] as a variation of Md and Mq ..................................................................................................
.....................................................................................................................................................184
198
CHAPTER 1
INTRODUCTION
1.1 Introduction
Basically there are two broad categories of ac generators; they are the
synchronous machines that generate output voltage; the frequency of the output voltage is
proportional to the rotational speed of the rotating parts. Induction generators are
primarily induction motors (machines) that are driven above their synchronous speed.
They cannot generate power on their own without some form of reactive power for
excitation which has to be provided through the stator winding. When the induction
machine is driven by an external prime mover, the flux in the rotor induces a small
voltage in the stator windings. The stator-winding voltage has a leading current if a
suitable source of reactive power is connected to the terminal. The source of reactive
induced voltages and currents would continue to rise, but the magnetic saturation in the
machine limits them. This results in a steady-state operating condition of the induction
squirrel-cage motor of the same output, needing neither field windings, exciter nor
1
199
automatic regulator. It has the disadvantages of needing supply to obtain the magnetizing
current and it generates at a leading power factor, requiring its magnetizing current to be
supplied by other synchronous generators or from the mains. This limitation can be
Induction generator applications in power systems have been studied [1]. The use
for excitation is supplied by the synchronous machine in the system while for a stand-
alone case the reactive power may be provided by a static VAR source such as capacitors
or an inverter.
electromotive force being induced in the machine windings. The induced voltages and
currents would continue to rise, but for the magnetic saturation in the machine that results
unit cost, brushless rotor (squirrel cage construction), absence of a separate dc source,
is its inability to control the supplied voltage and frequency under varying load
conditions.
variable speed wind turbine is discussed in [2]. Excitation control was obtained by
employing a capacitor and a thyristor-controlled inductor. The wind speed cube law was
used in loading the induction machine for maximizing energy conversion. Performance
characteristic of the generation scheme was evaluated over a wide speed range. The
harmonic analysis of the scheme shows that the harmonics generated by the converters
cage induction machine with self-excited capacitors. The self-excited induction generator
with controlled rectifier allows wide changes in wind-turbine speed with optimum
generating power set at all speeds by rectifier delay-angle control. A constant output dc
voltage can be maintained and the generator always operates in the low slip region.
Newton-Raphson method was used to predict the steady-state behavior of capacitor self-
excited induction generators. The values of the saturated magnetizing reactance and
ouput frequency of the given capacitance, speed, and load were also identified.
201
been examined in [5]. An analytical method was proposed to find the minimum
load conditions. The calculated values of the capacitance were compared with the
experimental results for some test machines. The influence of the load impedance, load
power factor, and machine speed on the value of minimum capacitor were also discussed.
generator when the excitation capacitor is connected to one winding and the load is
connected to the other is examined in [6]. A nonlinear analytical equation in terms of the
magnetization reactance and the operating frequency was developed. The resulting
equation was solved using Newton-Raphson numerical technique for the airgap voltage
and other performance indices. The predicted performance of the system is validated with
experimental results.
regulating features. The generator was designed with two uniquely designed stator
respectively. It employs a standard die-cast squirrel cage rotor. The features, advantages,
and theoretical concepts of the system were highlighted and a detailed experimental result
was presented.
based on the principles of harmonic balance were examined in [9]. The effects of
saturation of the magnetizing flux linkages and core loss of the generator were included
202
which supplies an isolated resistive load, was reported in [10]. The equivalent circuit
model was developed neglecting the magnetizing reactance in the negative sequence
circuits. Experimental results verified the equivalent circuit model and the assumptions
parallel connected load was discussed in [11]. The dependency of the load impedance on
the generator self-excitation was explored using the determinant of the generator steady-
state equation. It was concluded that for self excitation, a minimum amount of airgap flux
linkages is required. Maximum load impedance was specified for rotor speed and
excitation capacitors when load is connected in series with the main winding. If the load
connected in parallel with the capacitor, a minimum load impedance was required for
self-excitation.
short-shunt, and long-shunt, on the steady-state and dynamic performance of the single-
delivers lower output power at reduced output voltage while the short-shunt generator
system has load voltage characteristics and output power profile that can outperform the
same generator with the shunt excitation connection. Also the small-signal analysis
showed that the generator with either short-shunt or long-shunt excitation connections
overload. Also, a generator with either of these excitations quickly recovers when
overload is relieved.
The capacitor self-excited induction generator has limited applications due to its
inability to control the load voltage and frequency under varying load conditions. To cope
with the varying load and/or speed variation of the induction generator, the voltage can
characteristic and the stator and rotor resistances of the machines. Experimental results
The feasibility of the proposed scheme with naturally commutated thyristors was verified
units, which eliminate the problems of voltage and frequency variations inherent in self-
excited induction generator machines. Theoretical and experimental results are provided
showing that the self-excited induction generator can operate in the linear region of the
magnetization curve while feeding a variable dc load at constant voltage. It was also
shown that the unit could be used to feed controllable power into an existing a.c network
induction generator operation including both voltage and current inverter systems were
reviewed and compared. The equivalent circuit for each mode was presented and its use
to model steady-state performance was shown. The properties of regeneration and self-
excitation were derived to illustrate the similarities and differences between the various
modes. The experimental results were supported by the validity of the models used in the
study.
The main objective of this research is to present the dynamic and steady-state
which can be driven by wind or diesel engine. The proposed scheme, shown in Figure
1.1, has the advantage of ensuring a constant and regulated load voltage. In addition the
mathematical model developed for the single-phase induction generator scheme with
battery-PWM inverter will be useful in the prediction of the transient, dynamic, and
steady-state performance. The load voltage and frequency of the isolated single-phase
205
Vd Single
+ PWM phase Single
Battery Cd Induction phase
Inverter
- Generator Load
width modulated (PWM) dc/ac inverter. A battery feeding the single-phase inverter
provides real power to the load when the generator output is less than the real power
required by the load. If the generator delivers more real power than what is required by
the load, the battery through the bi-directional single-phase PWM inverter absorbs the
excess real power. The inverter augmented with a shunt capacitor connected across the
ii. As power sources (hybrid power generation) for isolated systems and for
which provides a constant voltage and frequency at the terminals, was investigated in
[17,32]. The system shows an induction generator to generate electric power and to filter
the current harmonics; a PWM inverter to provide the excitation and set the desired
frequency and an energy storage dc battery to allow a bi-directional power flow. The
[17,32], the inverter induces current and voltage harmonics that are directly transferred to
the load. Unlike the system proposed in [17,32] the scheme presented here does not
require an additional filter to remove current and voltage harmonics since the auxiliary
winding which is connected to the inverter is only magnetically connected to the main
winding acts as a filter. The mathematical model developed in this present research work
accurately predicts the experimental results unlike the model system proposed in [17,32].
The single-phase induction machine used for this work is rated at 1 h.p. The
operation of the system with battery, PWM (pulse width modulation) inverter, and
Experiments were performed with the generator system feeding impedance and
motor loads. The condition for self-excitation of the system without battery was also
studied.
Simulation of the system with impedance and motor loads was carried out
including the influence of variation of load and generator speed. MATLAB/Simulink [18]
207
developed.
The steady-state model of the system was developed using harmonic balance
technique [19]. The computation was carried out using MATHCAD [20], MATLAB, and
MATHEMATICAL [21].
Chapter 2 discusses the theory behind the single-phase pulse width modulation
detail in chapter 3. The various steps in arriving at the parameters of the single-phase
with an impedance load. The mathematical steps taken to arrive at the mathematical
model of the system are discussed. The dynamic mathematical model is simulated and
results are also discussed in this chapter. The dynamic performance and the influence of
single-phase induction motor load is set forth in chapter 5. The mathematical model
results are explained. The dynamic mathematical model of the generator system
connected to a single-phase induction motor (SPIM) load is simulated and the results are
208
without battery is the subject of discussion in chapter 6. The steps involved at arriving at
the condition of self-excitation are shown. The results of the numerical computation are
also discussed.
Chapter 7 includes conclusions and suggestions for further work on the system.
209
CHAPTER 2
2.1 Introduction
stiff dc voltage such as a battery or a rectifier feeds the voltage-source inverter. The filter
capacitor across the inverter input terminals provide a constant dc link voltage. The
phase-controlled thyristor feeds the inverter with a regulated current through a large
series inductor. Hence, the load current rather than load voltage is controlled.
or H-bridge, and push-pull transformer center tap configurations. The single-phase unit
supplies, uninterruptible power supplies (UPS), induction heating, and static var
generators.
In this chapter, different control schemes for single-phase PWM inverter will be
12
210
The PWM inverters have constant input dc voltage that is essentially constant in
magnitude, such as in the scheme in Figure 2.1. The battery or rectifier provides the dc
supply to the inverter. The inverter is used to control the fundamental voltage magnitude
and the frequency of the ac output voltages. This can be achieved by pulse-width
modulation of the inverter switches. Hence such inverters are called PWM inverters.
extensively reviewed in the literature [23-25]. The SPWM principle for a half-bridge
carrier wave. The point of intersection determines the switching points of the inverter
power devices. The inverter switching frequency is determined by the frequency of the
triangular waveform. The inverter switching frequency and amplitude are normally kept
Vd
Battery a.c.
+
or Cd Inverter Output
Rectifier Voltage
-
constant. The triangular waveform, vt, is at a switching or carrier frequency ft, which
determines the frequency with which the inverter switches are switched. The modulating
or control signal vm is used to modulate the switch duty ratio and has a frequency fm,
which is the desired fundamental frequency of the inverter voltage output. The amplitude
V̂m
ma = (2.1)
V̂t
where V̂m is the peak amplitude of the control signal. The amplitude V̂t of the triangular
(a)
(b)
Figure 2.2. Half-bridge PWM Inverter Waveforms. (a) Generation Method, (b) inverter
output voltage.
212
TA+ DA+
+
Vd
2 C
-
vo
Vd
+
Vd C TA-
DA-
2 -
ft
mf = . (2.2)
fm
In the inverter of Figure 2.3, the switches TA+ and TA- are controlled based on the
comparison of control signal, vm and triangular signal, vt, and the following output
Vd
v m > vt , TA+ is on, vo = (2.3)
2
and
Vd
vm < vt , TA − is on, vo = − . (2.4)
2
The switches on the same leg are protected against turning on at the same time.
This enables the output voltage, vo, to fluctuate between Vd/2 and - Vd/2. vo is shown in
Figure 2.2 b.
213
A single-phase inverter is shown in Figure 2.4. This inverter, also known as the
2.3. The full-bridge inverter can produce an output power twice that of the half-bridge
inverter with the same input voltage. Different PWM switching schemes will now be
discussed.
2.2.2.1 PWM with bipolar voltage switching. In this scheme the diagonally
opposite transistors (TA+, TB-) and (TA-, TB+) form the two legs in Figure 2.4 and are
switched as switch-pairs 1 and 2, respectively. The output of leg A is equal and opposite
to the output of leg B. The output voltage is determined by comparing the control signal,
vm, and the triangular signal, vt.. This is illustrated in Figure 2.5a.
DB+
TA+ TB+
+
Vd DA+
2 C
- A
vo
Vd o
B
+
Vd C TA-
DB-
TB-
2 - DA-
Vd Vd
vm > vt , T A + is on ⇒ v Ao = and TB − is on ⇒ v Bo = − ; (2.5a)
2 2
Vd V
v m < v t , T A− is on ⇒ v Ao = − and TB + is on ⇒ v Bo = d ; (2.5b)
2 2
hence
v Bo (t ) = − v Ao (t ) . (2.6)
(a)
(b)
(c)
Figure 2.5. PWM with bipolar voltage switching. (a) Generation Method, (b) inverter
output voltage, (c) inverter switching function.
215
The output voltage can then be expressed as shown in Equation 2.7 with a waveform
vo (t ) = v Ao (t ) − v Bo (t ) = 2v Ao (t ) (2.7)
The inverter switching function, which is similar to the inverter output voltage is shown
in Figure 2.5c.
In Figure 2.5, the output voltage vo produced by the modulating signal, vm and the
carrier signal, vt can be expressed by a complex double Fourier series. The output voltage
in Figure 2.5 is determined using the phase relationship between the modulating and
carrier signal shown in Figure 2.6 and 2.7. The peak value and angular frequency of the
1 vt vm
0
β1 π β2 2π
θ1 θ2 ωmt
a1
a2
-1
(a) Modulating and carrier signal waveforms, 0 ≤ ωm t + φ < π .
vo
Vd
0 π 2π
θ1 θ2 ω mt
-Vd
Figure 2.6. Phase relationship of output voltage to modulating and carrier signals for
bipolar voltage scheme. Condition, 0 ≤ ωm t + φ < π .
216
1 vt
a4
a3
θ3 π θ4 2π
0 β3 β4 ωmt
-1 vm
θ3 π θ4 2π
0
ω mt
-Vd
vo
carrier signal are 1 and ωt, respectively. The modulating signal is expressed as
The output voltage of Figure 2.5 can be expressed by complex double Fourier
∞ ∞
vo = ∑ ∑ K knexp( j{kx + ny}) (2.8)
k = - ∞ n = -∞
where x = ωmt + φ and y = ωt t . Equation 2.8 expressing the frequency spectrum of the
PWM waveform comprises the angular frequencies of n (integer) times ωt and the
217
Vd ∫0
(
∫0 ∫θ1 ∫θ2 )
πe − jkx − θ1 e − jny dy + θ 2 e − jny dy − 2 πe − jny dy dx
( )
K kn = 2 (2.9)
4π + 2 π e − jkx θ3 e − jny dy − θ4 e − jny dy + 2 π e − jny dy dx
∫π ∫0 ∫θ3 ∫θ 4
Angles θ1 and θ2 are the intersection points of vm and vt when modulating signal
v m = m a sin( ω m t + φ ) is greater than zero; and angles θ3 and θ4 are similar intersection
points of vm and vt when the modulating signal v m = m a sin( ω m t + φ ) is less than zero.
π π
θ1 + a1 = , ⇒ θ1 = − a1
2 2
1 ma sin x
tan β1 = , also tan β1 =
π a1
2
ma sin x 1 π
∴ = , ⇒ a1 = ma sin x
a1 π 2
2
Hence
π
θ1 = ( 1 − ma sin x ) . (2.10a)
2
Similarly, θ2, θ3, and θ4 are determined; their values are given as:
π
θ2 = ( 3 + ma sin x )
2
π
θ3 = ( 1 − ma sin x ) (2.10b)
2
218
π
θ4 = ( 3 + ma sin x )
2
Equation 2.8 will now be separated into its respective frequency components.
2V d 2V d
v 00 = ma and K 00 = ma
π π
Kk0 =
Vd
4π 2
{∫ e
π − jkx
0
(2θ 2 − 2θ1 − 2π) dx + ∫π2 π e − jkx (2θ 3 + 2π − 2θ 4 ) dx }. (2.11)
But
Kk0 =
Vd
4π
{∫ e
π − jkx
0
(2m a sin x ) dx + ∫π2 π e − jkx (− 2m a sin x ) dx } (2.13)
where
Vd m a 2
− (k = 2, 4, 6, L)
K ko = π k 2 − 1 (2.14)
0 (k : odd number )
The output voltage for the harmonic components, vko is obtained by substituting
Equation 2.13 into Equation 2.8. The output voltage by performing real Fourier
(k + 2 )
(− 1) 2
4m aVd sin x (k = 2 , 4, 6, L)
v ko = k 2 − 1 π (2.15)
0 (k : odd number )
In Equation 2.15, the fundamental component of the signal is zero and the
K kn =
4π
Vd
2
{∫ 2 π − jkx
0
e (− ∫
θ1 − jny
0
e dy
θ
1
2π
2
) }
+ ∫θ 2 e − jny dy − ∫θ e − jny dy dx
K kn =
Vd
j 4π n
{∫
2
2 π − jkx
0
e (2e − jnθ1
) }
− 2e − jnθ 2 + e − j 2 nπ − 1 dx (2.16)
The value of Kkn from Equation 2.17 when n is an even number is given as
n
2Vd 2 π − jkx 2 jnπma sin x
1 1
− jnπm a sin x
K kn = (− 1)
2
− − dx
2 ∫0
e e e 2 1 (2.18)
j 4π n
n
(− 1) 2 2Vd J ma nπ (k : odd number )
K kn = k (2.19)
j πn 2
0 (k : even number )
where
m nπ 1 π m nπ
J k a = ∫0 sin kx sin a sin x dx
2 π 2
220
m nπ
The Jk is the kth-order Bessel function of the first kind. J k a is a function whose
2
The value of Kkn from Equation 2.17 when n is an odd number is given as:
(n +1)
2Vd 2 π − jkx 2 jnπma sin x − jnπm a sin x
1 1
K kn = (− 1)
2
2 ∫0
e e +e 2 dx (2.20)
4π n
(n +1)
m nπ
= (− 1) (k : odd number )
2 2Vd
K kn Jk a (2.21)
πn 2
0 (k : even number )
where
m nπ 1 π a m nπ
Jk a = ∫0 cos kx cos 2 sin x dx
2 π
(k +1)
2V ∞ 4 m aVd
v o = d m a + ∑ (− 1) sin(kω m t + kφ)
2
π ( k 2 − 1)π
k = 2 ,4L
(n +1) m a nπ ∞
m nπ
∞ 4Vd o
J cos n ω t + ∑ Jk a
∑ (− 1)
2 t
+ 2 k = 2 ,4 2
nπ
n =1,3L
× [cos ((nω t + k ω m ) t + k φ ) + cos ((nω t − k ω m )t + k φ )]
n
∞ 4Vd ∞ m nπ
∑ (− 1) ∑ J k a [ sin((nωt + kωm )t + kφ) − sin((nωt − kω m )t + kφ) ]
2
+
n = 2 ,4L nπ k =1,3 2
(2.22)
The first term in Equation 2.22 is the dc component of the output voltage. The second
expresses the harmonic components of even terms of the signal. The third and fourth
terms give the sideband components when n is an odd number and an even number,
respectively.
221
Vo = MVd (2.23)
where
2ma -1
M= sin α + α 1 − α , ma > 1
2
π
α= 1 .
ma
2.2.2.2 PWM with unipolar voltage switching. In this scheme, the diagonally
opposite transistors (TA+, TB-) and (TA-, TB+) form the two legs of the full-bridge inverter
Figure 2.4 and are not switched simultaneously. The legs A and B of the full-bridge are
controlled by comparing triangular signal, vt. with control signal, vm and -vm, respectively.
The comparison of vm and vt provides logic signals to control the switches in leg
A:
v m > v t : T A+ on and v AN = V d
and (2.24)
v m < v t : T A − on and v AN = 0 .
The output of the inverter leg A and leg B with respect to the negative dc bus N is shown
in Figures 2.8b and 2.8c respectively. For controlling the leg B transistors, (-vm) is
− v m > v t : T B+ on and v BN = V d
and (2.25)
v m < v t : T B − on and v BN = 0 .
222
(a)
(b)
(c)
(d)
Figure 2.8. Single-Phase PWM with unipolar voltage switching. (a) Generation Method,
(b) leg A output voltage, (c) leg B output voltage, (d) inverter output voltage.
223
Table 2.1. Switching state of the unipolar PWM and the corresponding voltage levels.
ON - - ON Vd 0 Vd
- ON ON - 0 Vd -Vd
ON - ON - Vd Vd 0
- ON - ON 0 0 0
Table 2.1 shows the switching state of the unipolar PWM and the corresponding
voltage levels. It should be noticed that when both the upper switches are on, the output
voltage is zero. The output current circulates in a loop through (TA+ and DB+) or (DA+ and
TB+) depending on the direction of inverter output current, io. During this interval, the
inverter input current id is zero. A similar condition occurs when both bottom switches
In this type of PWM scheme, when a switching occurs, the output voltage
changes between 0 and +Vd or between 0 and -Vd voltage levels. For this reason, this type
of PWM scheme is called the pulse width modulation with a unipolar voltage switching,
as opposed to the PWM with bipolar (between +Vd and -Vd). This scheme has the
harmonics are concerned, compared to the bipolar-voltage switching scheme. Also the
voltage jumps in the output at each switching are reduced to Vd, as compared to 2Vd in
2.2.2.3 Modified PWM bipolar voltage switching scheme. In this scheme, the
diagonally opposite transistors (TA+, TB-) and (TA-, TB+) form the two legs of the full
224
bridge inverter Figure 2.4 and are not switched simultaneously. The output voltage is
determined by comparing the control signal, vm, and the triangular signal, vt.. This is
illustrated in Figure 2.9a. The switching pattern for positive values of modulating signal,
vm is as follows
v m > v t , T A+ is on ⇒ v AN = Vd
v m < v t , T B + is on ⇒ v BN = Vd
(a)
(b)
Figure 2.9. PWM with modified bipolar voltage switching. (a) Generation Method, (b)
inverter output voltage.
225
The output voltage can then be expressed as shown in Equation 2.28 with a waveform
v o (t ) = v AN (t ) − v BN (t ) (2.28)
Table 2.2 shows the switching state of the modified bipolar PWM and the
corresponding voltage levels. It should be noticed that when both the upper switches are
on, the output voltage is zero. The output current circulates in a loop through (TA+ and
DB+) or (DA+ and TB+) depending on the direction of inverter output current, io. During
this interval, the inverter input current id is zero. A similar condition occurs when both
bottom switches TA- and TB- are on. In the modified bipolar PWM scheme, when a
switching occurs, the output voltage changes between 0 and +Vd or between 0 and -Vd
voltage levels.
Table 2.2. Switching state of the modified bipolar PWM and the corresponding voltage
levels.
TA+ TA- TB+ TB- vAN vBN Vo
ON - - ON Vd 0 Vd
- ON ON - 0 Vd -Vd
ON - ON - Vd Vd 0
- ON - ON 0 0 0
In Figure 2.9, the output voltage vo produced by the modulating signal, vm and the
carrier signal, vt can be expressed by a complex double Fourier series. The output voltage
in Figure 2.9 is determined using the phase relationship between the modulating and
carrier signal shown in Figure 2.10 and 2.11. The peak value and angular frequency of
226
the carrier signal are 1 and ωt, respectively. The modulating signal is expressed as
The output voltage of Figure 2.9 can be expressed by complex double Fourier
∞ ∞
vo = ∑ ∑ K kn exp( j{kx + ny}) (2.29)
k =- ∞ n = -∞
PWM waveform comprises the angular frequencies of n (integer) times ωt and the
1 vt vm
0
β1 π β2 2π
θ1 θ2 ωmt
a1
a2
-1
(a) Modulating and carrier signal waveforms, 0 ≤ ωm t + φ < π .
vo
Vd
0 π 2π
θ1 θ2 ωmt
1 vt
a4
a3
θ3 π θ4 2π
0 β3 β4 ωmt
-1 vm
θ3 π θ4 2π
0
ω mt
-Vd
vo
(b) Inverter output voltage, π ≤ ωm t + φ < 2π .
Figure 2.11. Phase relationship of output voltage to modulating and carrier signals for
modified bipolar voltage scheme. Condition, π ≤ ωm t + φ < 2π .
K kn =
Vd
4π 2
{∫ e (∫
π − jkx θ 2
0 θ1
) 2π
e − jny dy dx − ∫π e − jkx (∫θ 3 − jny
0
e dy
2π
) }
+ ∫θ e − jny dy dx
4
(2.30)
Angles θ1 and θ2 are the intersection points of vm and vt when the modulating
signal v m = m a sin( ω m t + φ ) is greater than zero; and angles θ3 and θ4 are the
π π
θ1 + a1 = , ⇒ θ1 = − a1
2 2
228
1 ma sin x
tan β1 = , also tan β1 =
π a1
2
ma sin x 1 π
∴ = , ⇒ a1 = ma sin x
a1 π 2
2
Hence
π
θ1 = ( 1 − ma sin x ) . (2.31a)
2
Similarly, θ2, θ3, and θ4 are determined; their values are given as
π
θ2 = ( 3 + ma sin x )
2
π
θ3 = ( 1 − ma sin x ) (2.31b)
2
π
θ4 = ( 3 + ma sin x )
2
Equation 2.29 will now be separated into its respective frequency components.
n = 0:
Kk 0 =
Vd
4π 2
{∫ e
π − jkx
0
(θ2 − θ1 ) dx − ∫π2 π e − jkx (θ3 + 2π − θ4 ) dx } (2.32)
But
Kk 0 =
Vd
4π 2
{∫ e
π − jkx
0
(π(1 + ma sin x )) dx − ∫π2 π e − jkx (π(1 − ma sin x )) dx } (2.34)
where
2 π − jkx − jma π (k = 1)
∫0 e (ma sin x ) dx = (2.35a)
0 (k ≠ 1)
π − jkx 2 π − jkx − j
4
(k : odd number )
∫0
e dx − ∫ π
e dx = k (2.35b)
0 (k : even number )
1 m
− jVd + a (k = 1)
π 4
V
K ko = − j d (k : odd number ) (2.36)
πk
0 (k : even number )
The output voltage for the harmonic component, vko can be obtained by
substituting Equation 2.34 into Equation 2.29. The output voltage by performing real
MVd sin x (k = 1)
vko = 2Vd (2.37)
kπ sin kx (k = 3, 5, 7 L)
where
2 m
M = + a
π 2
K kn =
Vd
4π 2
{∫ 0
e θ1
(∫
2 π − jkx θ 2
) }
e − jny dy dx
K kn =
Vd
j 4π n
{∫
2
2 π − jkx
0
e (e − jnθ1
) }
− e − jnθ 2 dx (2.38)
2 π − jkx − 2 jnπ 2 jnπma sin x − jnπ − jnπm a sin x
1 1 3 1
Vd
K kn = 2 ∫0
e e e −e 2 e 2 dx (2.39)
j 4π n
The value of Kkn from Equation 2.39 when n is an even number is given as
n
2 π − jkx 2 jnπma sin x − jnπm a sin x
1 1
K kn = (− 1)
2 Vd
2 ∫0
e e −e 2 dx (2.40)
j 4π n
n
m a nπ
= (− 1) jπn J k 2 (k : odd number )
2 Vd
K kn (2.41)
0 (k : even number )
where
m nπ 1 π m nπ
J k a = ∫0 sin kx sin a sin x dx
2 π 2
m nπ
The Jk is the kth-order Bessel function of the first kind. J k a is a function whose
2
The value of Kkn from Equation 2.39 when n is an odd number is given as
(n +1)
Vd 2 π − jkx 2 jnπma sin x − jnπm a sin x
1 1
K kn = (− 1)
2
+ dx
∫ e
e e 2
(2.42)
4π 2 n 0
231
(n +1)
m nπ
= (− 1) (k : odd number )
2 Vd
K kn Jk a (2.43)
πn 2
0 (k : even number )
where
m nπ 1 π m nπ
a
Jk a = ∫0 cos kx cos 2 sin x dx
2 π
∞
2Vd
vo = MVd sin(ωm t + φ) + ∑ sin(kωm t + kφ)
k = 3,5L kπ
(n +1) ma nπ ∞
m nπ
∞
2Vd J o cos nωt t + ∑ J k a
∑ (− 1)
2
+ 2 k = 2 ,4 2
nπ
n =1,3L
× [cos ((nωt + kωm )t + kφ) + cos ((nωt − kωm )t + kφ)]
n
∞
2Vd ∞ m nπ
∑ (− 1) ∑ J k a [ sin((nωt + kωm )t + kφ) − sin((nωt − kωm )t + kφ) ]
2
+
n = 2 ,4L nπ k =1,3 2
(2.44)
The first term in Equation 2.44 is the fundamental component of the output voltage. The
second term expresses the harmonic components of odd term of the fundamental signal
and is a term not related to the modulation index. The third and fourth terms give the
Vo = MVd (2.45)
where
4
M = 0 .5 × m a + , ma ≤ 1 .
π
232
The single-phase PWM inverter can be switched with different voltage switching
schemes as discussed. Basically, the transistor used in the single-phase inverter has to be
driven into the saturation region to act as a switch. The base signal used in driving the
IGBT and MOSFET Driver). The base driver has the following features [26]: CMOS
compatible inputs, interlock circuit, short circuit monitoring and switch off, supply
transformer, error latch and monitoring output signal, and internal isolated power supply.
The supply to base drive board is +15 V. The input into the base driver (4 of 6) is
obtained with operational amplifier and resistors. The pulse width modulation is
The PWM can be obtained by using op amp (operational amplifier). The pulse
From the theory in section 2.2.2.1, the PWM bipolar-voltage switching method is
explained. The PWM with bipolar voltage switch scheme is obtained with operational
2.12. The comparator circuit compares a signal voltage on one input of an op amp with
known voltage called reference voltage on the other input. Its output may be (+) or (-)
saturation voltage, depending on which input is larger. Figure 2.13 shows the dual low
noise JFET input op-amp used (ECG 858M). In achieving the bipolar voltage switching
233
R1 + VCC
Vsine - Vo
Vtri +
R1 - VCC
R2
OUTPUT 1 8 + VCC
INV INPUT 2 - 7 OUTPUT
NON-INV INPUT 3 +
- 6 INV INPUT
scheme, the transistors TA+ (TOP 1) and T B- (BOT 2) are turned on from the output of
Figure 2.14.
to the (+) input, and the modulating signal (sinusoidal waveform, 0V - 20V) is applied to
the (-) input. The circuit arrangement of Figure 2.14 is called the inverting comparator.
When Vsine is greater than Vtri, the output voltage Vo is at -15V because the voltage at the
(+) input is lesser than that at the (-) input. On the other hand, when Vsine is less than Vtri,
the (-) input becomes negative with respect to the (+) input, and Vo goes to +15V. Hence
Vo changes from one saturation level to another whenever Vsine > Vtri or Vsine < Vtri.
234
R2
R1 + 15V
Vsine - Vo TOP 1
Vtri + BOT 2
R1
- 15V
R2
R2
R1 + 15V
Vtri - Vo BOT 1
Vsine TOP 2
+
R1
- 15V
R2
The other set of transistors TA- (BOT 1) and T B+ (TOP 2) are switched using
Figure 2.15. This is done using the bipolar voltage-switching scheme also. The fixed
reference voltage (triangular waveform) of 8 V peak to peak is applied to the (-) input,
and the modulating signal (sinusoidal waveform, 0V - 20V) is applied to the (+) input.
When Vsine is less than Vtri, the output voltage Vo is at -15V because the voltage at the (-)
input is higher than that at the (+) input. Conversely, when Vsine is greater than Vtri, the
(+) input becomes positive with respect to the (-) input, and Vo goes to +15V. Hence Vo
changes from one saturation level to another whenever Vsine > Vtri or Vsine < Vtri.
235
The H-bridge inverter shown in Figure 2.4 is implemented using two sets of insulated
gate bipolar transistor (IGBT). Each set of the international rectifiers (IRGTI050U06) is a
half-bridge that consists of two transistors. The choice of IGBT for this power application
was based on the facts that it has the following features [5]: rugged design, simple gate-
switching-loss rating. The outputs from the base drive were connected appropriately to
the collector, emitter, and gate of the IGBT. The interlock circuit on the base driver
ensures the IGBTs from the same leg do not turn on simultaneously.
236
CHAPTER 3
3.1 Introduction
examined. First, the derivation of the single-phase induction motor (SPIM) equation is
carried out. In simplifying the SPIM state-variables equations, two sets of transformation
are used: the stationary reference frame and the turn ratio transformation. Next, a brief
The harmonic balance technique is then used to develop a steady-state model for
the SPIM. The parameters of the SPIM are determined with the resulting steady-state
equation.
Figure 3.1. The stator or main windings are nonidentical sinusoidally distributed
turns with resistance ra. The auxiliary bs winding has Nd equivalent turns with resistance
rb. The rotor windings may be considered as two identical sinusoidally distributed
windings
39
237
bs axis
br axis as’
ωr
ar’ ar axis
br θr
bs
bs’
as axis
br’
ar
as
arranged in space quadrature. Each rotor winding has Nr equivalent turns with resistance
rr.
where ra, rb, and rr are the resistances of the a-phase stator winding, b-phase stator
winding, and rotor winding, respectively. The p denotes differentiation with respect to
λ abs L s L sr i abs
λ = (L )T L r i abr
(3.5)
abr sr
where
238
L Lasbs Llas 0
L s = asas +
Lbsbs 0 Llbs
(3.6)
Lbsas
L Larbr Llr 0
L r = arar +
Lbrbr 0 Llr
(3.7)
Lbrar
Lasar Lasbr
L sr =
Lbsbr
(3.8)
Lbsar
where Ls is the self-inductance for the stator winding, Lr is the self-inductance for the
Winding function principles [28] can be used in determining the self- and mutual
inductance matrix in Equations 3.6 - 3.8. The inductance between any windings “i” and
µ o rl 2π
Lij =
g 0
∫ N i ( φ )N j ( φ )dφ . (3.9)
The average air-gap radius is r, the effective motor stack length is l, and the gap length is
represented by g. The angle φ defines the angular position along the stator inner diameter,
while the angular position of the rotor with respect to the stator reference frame is φr.
The winding functions of windings “i” and “j” are given, respectively, as Ni (φ) and Nj
(φ). In order to obtain closed-form inductance equations, the winding functions are
The fundamental component of the winding functions for the stator windings are
given in Equation 3.10. The fundamental component of the winding functions for the
The self- and mutual inductances can be found using Equation 3.9-3.11. This
µ o rl 2π
Lasas = ∫ ( N A cos( φ ))2 dφ = µ o rl πN A2 (3.12)
g 0 g
µ o rl 2 π
Lasbs = ∫ N A cos( φ ) × N B sin( φ )dφ = 0 (3.13)
g 0
µ o rl 2 π
Lbsas = ∫ N B sin( φ ) × N A cos( φ )dφ = 0 (3.14)
g 0
µ o rl 2π
Lbsbs = ∫ ( N B sin( φ ))2 dφ = µ o rl πN B2 (3.15)
g 0 g
µ o rl 2π
Larar = ∫ ( N r cos( φ − φr ))2 dφ = µ o rl πN r2 (3.16)
g 0 g
µ o rl 2 π
Larbr = ∫ N r cos( φ − φr ) × N r sin( φ − φr )dφ = 0 (3.17)
g 0
µ o rl 2 π
Lbrar = ∫ N r sin( φ − φr ) × N r cos( φ − φr )dφ = 0 (3.18)
g 0
µ o rl 2π
Lbrbr = ∫ ( N r sin( φ − φr ))2 dφ = µ o rl πN r2 (3.19)
g 0 g
µ o rl 2 π µ rl
Lasar = ∫ N A cos( φ ) × N r cos( φ − φr )dφ = o πN A N r cos( φr ) (3.20)
g 0 g
µ o rl 2 π µ rl
Lasbr = ∫ N A cos( φ ) × N r sin( φ − φr )dφ = − o πN A N r sin( φr ) (3.21)
g 0 g
µ o rl 2 π µ rl
Lbsar = ∫ N B sin( φ ) × N r cos( φ − φr )dφ = o πN B N r sin( φr ) (3.22)
g 0 g
240
µ o rl 2 π µ rl
Lbsbr = ∫ N B sin( φ ) × N r sin( φ − φr )dφ = o πN B N r cos( φr ) (3.23)
g 0 g
µ o rl
where k = π.
g
In Equations 3.24 and 3.25, the leakage inductances of the a-phase and b-phase
stator windings are Llas and Llbs, respectively, while those of the rotor windings is Llr.
rotor variables to the arbitrary reference frame are necessary in simplifying the SPIM
f as
f abs = (3.27)
f bs
241
where f may be a voltage (v), current (i), or flux linkages (λ). If θ is the angular position
of an arbitrary reference frame then the transformation of the stator variables to arbitrary
where
f qs
f qds = (3.29)
f ds
and
r 0
where rab = a
0 rb
ra + rb ra − rb ra − rb
vqs 2 + 2 cos 2θ 2
sin 2θ iqs λ ds λ
v = ra − rb ra + rb rb − ra + ω + p qs (3.33)
ds sin 2θ + cos 2θ ids − λ qs − λ ds
2 2 2
where ω = pθ.
resistance. The equation can be simplified using the stationary reference frame, ω = 0
f ar
f abr = . (3.35)
f br
where
f qr
f qdr = (3.37)
f dr
and
vqr iqr λ dr λ qr
v = rr i + (ω − ωr )− λ + p − λ (3.41)
dr dr qr dr
where ω = pθ.
The rotor voltage Equation 3.41 using stationary reference frame, ω = 0 becomes
vqr iqr λ dr λ qr
v = rr i - ωr − λ + p − λ . (3.42)
dr dr qr dr
243
The flux linkage Equation 3.5 is transformed to the stationary reference frame as follows
1 0
(K 2s )-1 =
1 0
K 2s = , , (3.44a)
0 − 1 0 − 1
Therefore,
K 2 s L s (K 2 s )−1 = L s (3.45)
K 2 r L r (K 2 r )−1 = L r (3.46)
N N 0
K 2 s L sr (K 2 r )−1 = K 2 r (L sr )T (K 2 s )−1 = k × A r
N B N r
. (3.47)
0
where
N N 0 Lqr 0
L12 = k × A r =
N B N r 0 Ldr
. (3.49c)
0
244
In Equations 3.49a - 3.49b, the leakage inductances of the stator main and
auxiliary windings are Llqs and Llds, respectively, while those of the effective two
and all d variables to the bs winding with Nd effective turns. If all the q variables are
referred to the winding with Nq effective turns (as winding) and all d variables are
referred to the winding with Nd effective turns (bs winding), the voltage equation
becomes
Nq
′ = rqr
vqr ′ iqr
′ − ωr λ′dr + pλ′qr (3.52)
Nd
Nd
′ = rdr
vdr ′ idr
′ + ωr λ′qr + pλ′dr (3.53)
Nq
where
where Lqs = Llqs + Lmq , Lds = Llds + Lmd , Lqr = Llqr + Lmq and Ldr = Lldr + Lmd . In
2
Nq N N N N
′ =
which vqr ′ = r iqr , vdr
vqr , iqr ′ = d vdr , idr ′ = q rr ,
′ = r idr , rqr
Nr Nq Nr Nd Nr
245
2 2 2
N N N Nq N
′ = d rr , Lldr
′ = q Llqr , rdr
Llqr ′ = d Lldr , Lmq = Lqsr , Lmd = d Ldsr .
Nr Nr Nr Nr Nr
Since the source voltages in the rotor circuit are zero, the SPIM voltage equation in the q-
′ iqr
0 = rqr ′ − N qd ωr λ ′dr + pλ ′qr (3.60)
′ idr
0 = rdr ′ + N dq ωr λ ′qr + pλ ′dr (3.61)
where,
Nq Nd
N qd = and N dq = .
Nd Nq
Wf =
1
(i abs )T (Ls − Llabs I ) iabs + (iabs )T (Lsr ) iabr + 1 (i abr )T (Lr − Llabr I ) iabr (3.62)
2 2
where Te is the electromagnetic torque positive for motor action (torque output) and θrm
is the actual angular displacement of the rotor. The flux linkages, currents, and Wf
are all expressed as functions of the electrical angular displacement θr. Since
246
P
θ r = θ rm (3.64)
2
where P is the number of poles in the machine, then Equation (3.63) becomes
2
dWm = −Te dθ r . (3.65)
P
( )
Te i j ,θr =
(
P ∂W f i j ,θr ) . (3.66)
2 ∂θr
Since Ls and Lr are not functions of θr, substituting Wf from Equation 3.62 in Equation
P ∂
Te = (iabs )T [Lsr ] iabr . (3.67)
2 ∂θr
P
[ ]
Te = Lqr ias (− iar sin θ r − ibr cos θ r ) + Ldr ibs (iar cos θ r − ibr sin θ r ) . (3.68)
2
2
Te = J pωr + TL (3.69)
P
where J is the inertia of the rotor and the connected load. The load torque is TL; it is
P
[
Te = (K s )−1 i qds
2
] T ∂
∂θ r
[
[Lsr ] (K r )−1 i qdr ] (3.70c)
247
Te =
P
2
(
Lqr iqsidr − Ldr idsiqr . ) (3.72)
In terms of referred, substitute variables the expression for the torque becomes
Te =
P
2
(
N dq λ ′qr idr )
′ − N qd λ ′dr iqr . (3.73)
reasons:
(b) The technique is very efficient, hence it is used in power systems, electric
calculated.
248
Amcos(ωt + φ) (3.74a)
where the amplitude Am , the frequency ω, and the phase φ are real constants. The
amplitude or the peak is assumed positive. The frequency ω is measured in radians per
second. The period is T = 2π/ω in seconds. The sinusoidal function is of the form
Am cos(2πft + φ) (3.74b)
A = Am e jφ (3.75)
where A is the complex number called the phasor. If the frequency ω is known, then the
[ ] [ ]
Re Ae jωt = Re Am e j(ωt + φ ) = Am cos(ωt + φ ) (3.76)
The relationship between sinusoids (at frequency ω) and phasor can be given as
circuits in sinusoidal steady state are based on the following theorems [19].
Theorem 1 (Uniqueness). Two sinusoids are equal if and only if they are represented by
[ ] [ ]
Re Ae jωt = Re Be jωt ⇔ A = B . (3.79)
[ ] [
Re Ae jωt = Re Be jωt . ] (3.80)
Similarly, for to = π/2ω, exp jωto = exp j(π/2) = j and Re[Aj] = - Im[A]; hence Equation
(3.80) becomes
real coefficients) and is equal to the same linear combination of the phasors representing
[
x1 (t ) = Re A1e jωt ] and [
x2 (t ) = Re A2e jωt .]
Thus the phasor A1 represents the sinusoid x1 (t ) and the phasor A2 represents x2 (t ) .
[ ]
a1 x1 (t ) + a2 x2 (t ) = a1 Re A1e jωt + a2 Re A2e jωt . [ ] (3.83)
Now a1 and a2 are real numbers, hence for any complex numbers z1 and z2,
[ ] [ ]
a1 Re A1e jωt + a2 Re A2e jωt = Re (a1A1 + a2 A2 )e jωt [ ] (3.85)
[
a1 x1 (t ) + a2 x2 (t ) = Re (a1A1 + a2 A2 )e jωt . ] (3.86)
Symbolically,
[
Re jωAe jωt = ] d
dt
( [
Re Ae jωt ]) . (3.87)
d
Equation 3.87 indicates that the linear operators Re and commute:
dt
251
d
(
Re Ae jωt
dt
) = Re[jωAe jωt ] = dtd [Re(Ae jωt )] .
PROOF: By calculation. A = Amexp( j∠A ) .
d
[Re( A exp( jωt ))]
dt
=
d
[Re( Amexp j (ωt + ∠A ))] where Am is real
dt
=
d
[Amcos(ωt + ∠A )] since exp jx = cos x + j sin x
dt
= − Amωsin(ωt + ∠A )
Re[ A] × Re[B ] =
1
2
( [ ])
Re[ AB ] + Re AB* (3.88)
Re[ A] × Re[B ] =
1
2
( [ ]
Re A* B + Re[ AB ] ) (3.89)
[ ] [ ] [ ] [ ]
Vqs = Re Vqsse jωe t , Vds = Re Vdsse jωe t , λ qs = Re λ qss e jωet , λ ds = Re λ dsse jωe t ,
252
[ ] [
λ ′qr = Re λ ′qrr e jωe t , λ ′dr = Re λ ′drr e jωe t ], I qs [ ] [ ]
= Re I qsse jωe t , I ds = Re I dsse jωe t ,
I qr [′ e jω e t
′ = Re I qrr ] and ′ = Re I drr
I dr [
′ e jω e t ].
where Vdss, Vqss, λdss, λqss, λ'qrr, λ'drr, Iqss, Idss, I'qrr, and I'drr are complex peak quantities.
At steady state the coefficients of the state variables in Equations 3.58-3.61 are time
[ ] [ ] { [
Re Vdsse jωe t = rds Re I dsse jωe t + p Re λ dsse jωe t ]} ; (3.90a)
Similarly
[ ] [ ] { [
Re Vqsse jωe t = rqs Re I qsse jωe t + p Re λ qsse jωe t ]} (3.91a)
[ ] [ ] { [
′ e jωe t − N qd ωr Re λ ′drr e jωe t + p Re λ ′qrr e jωe t
′ Re I qrr
0 = rqr ]} (3.92a)
′ I qrr
0 = rqr ′ − N qd ωr λ ′drr + jωe λ ′qrr + pλ ′qrr (3.92b)
[ ] [ ] { [
′ e jωe t + N dq ωr Re λ ′qrr e jωe t + p Re λ ′drr e jωe t
′ Re I drr
0 = rdr ]} (3.93a)
′ I drr
0 = rdr ′ + N dq ωr λ ′qrr + jωeλ ′drr + pλ ′drr . (3.93b)
′
λ qss = Lqs I qss + Lmq I qrr (3.94)
′
λ dss = Lds I dss + Lmd I drr (3.95)
λ ′qrr = Lqr
′ I qrr
′ + Lmq I qss (3.96)
λ ′drr = Ldr
′ I drr
′ + Lmd I dss . (3.97)
253
[
Te = Re Te1 + Te2e j 2ωe t . ] (3.98)
Using harmonic balance technique the steady state electromagnetic torque can be found
[ ]
Re Te1 + Te2e j 2ωe t
[
Re Te1 + Te 2 e j 2 ωet ]
=
P N dq
2 2
( [
Re λ ′qrr I drr ]
′* + Re λ ′qrr I drr[
′ e j 2 ωet ] ) - N2 (Re[λ ′
qd
′*
drr I qrr ]+ Re[λ ′ ′ e
drr I qrr
j 2 ωet
] ).
(3.100)
4
(
Te1 = P N dq λ ′qrr I drr
′* − N qd λ ′drr I qrr
′* ) (3.101)
4
(
Te 2 = P N dq λ ′qrr I drr
′ − N qd λ ′drr I qrr
′ ) (3.102)
where Te1 and Te2 are the average and pulsating electromagnetic torque, respectively.
The equivalent circuit for the system using voltage equations (3.90b-3.93b) is
Core loss resistances Rmq and Rmd are added to the equivalent circuit in Figure 3.2
to account for core loss. The new equivalent circuit including core resistance is shown in
Figure 3.3. The steady-state equation the for new equivalent circuit is given in Equation2
3.103-3.110
Ndqωr λ’ qrr
rds Llds r’dr L’ldr
- +
Idss Lmd
Vdss I’drr
(a)
(b)
Figure 3.2. Equivalent circuit of single-phase induction machine (a) d-axis, (b) q-axis.
Ndqωr λ’ qrr
rds Llds r’dr L’ldr
- +
Idm
Idss
Vdss Lmd I’drr
Rmd
0 = N qd ωr λ ′drr − rqr
′ I qrr
′ − jωe λ ′qrr (3.105)
0 = − N dq ωr λ ′qrr − rdr
′ I drr
′ − jωe λ ′drr (3.106)
where the q-d flux linkages are defined in terms of the steady state q-d currents as
( ′ − I qm
λ qss = Llqs I qss + Lmq I qss + I qrr ) (3.107)
λ ′qrr = Llqr
′ I qrr (
′ + Lmq I qrr
′ + I qss − I qm ) (3.109)
The q-d flux linkages (3.107-3.110) are then substituted in the voltage equations (3.103-
′ I qrr
0 = rqr
′ − N qd ωr Lldr I drr
′ + Lmd Rmd
(I dss + I drr
′ )
′ I qrr
′ K
+ jωe Llqr
( Rmd + j ω e Lmd )
′ I drr
′ + N dq ω r Llqr I qrr
′ + Lmq Rmq
( ′
I qss + I qrr )
′ I drr
′ K
0 = rdr + jω e Lldr
(
Rmq + jωe Lmq )
(3.114)
(I dss + I drr
′ )
+ jωe Lmd Rmd .
(Rmd + jωe Lmd )
256
The main and auxiliary winding resistances were found by applying a dc voltage
across the terminals of the main and auxiliary windings. The values are main winding
The other parameters are estimated using the stand-still and synchronous-speed
tests. The stand-still test is carried out by holding the rotor down and by applying a
single-phase voltage across the auxiliary winding with the main winding opened. As the
voltage across the auxiliary winding is increased gradually, the auxiliary winding input
power, phase voltage, and current are measured along with the voltage across the main
winding. Also, the voltage source is applied across the main winding with the auxiliary
winding open. Main winding input power, phase voltage, and current, as well as the
voltage at the terminals of the auxiliary winding, are then measured as the main voltage is
increased.
The synchronous test is performed in addition to the stand-still test. This test is
accomplished by running the rotor at the synchronous speed. Then all other steps of the
test are carried out as was described in the stand-still test. The measured experimental
Table 3.1: Stand-still test: voltage applied to (a) the auxiliary winding and (b) the main
winding.
(a) (b)
V1A [V] I1A [A] P1A [W] V2M [V] V1M [V] I1M [A] P1M [W] V2A [V]
3.162 0.1662 0.264 0.192 3.79 0.331 0.585 0.025
6.346 0.32 0.9672 0.539 5.049 0.555 1.076 0.032
8.843 0.5438 2.474 0.685 7.23 1.052 3.188 0.044
11.98 0.8693 5.957 0.627 10.16 1.765 8.374 0.082
16.28 1.3409 13.77 0.597 13.71 2.687 19.07 0.12
20.6 1.8034 24.92 0.498 17.33 3.672 36.01 0.156
24.27 2.2 37.22 0.434 19.58 4.224 46.92 0.182
27.31 2.53 49.29 0.373 25.24 5.67 84.9 0.256
Notes: where V1A is the auxiliary winding voltage with voltage applied to the auxiliary
winding, I1A is the auxiliary winding current with voltage applied to the auxiliary
winding, P1A is the power to auxiliary winding power with voltage applied to the
auxiliary winding, V2M is the main winding voltage with voltage applied to the auxiliary
winding, V1M is the main winding voltage with voltage applied to the main winding,
I1M is the main winding current with voltage applied to the main winding, P1M is the
power to the main winding power with voltage applied to the main winding, and V2A is
the auxiliary winding voltage with voltage applied to the main winding,
relationship V1M= Vqs, I1M= Iqs, V1A= Vds, and I1A= Ids.
From the data in Table 3.1b, the main winding rotor resistance, rqr can be found.
Table 3.2: Synchronous test: voltage applied to (a) the auxiliary winding and (b) the main
winding.
(a) (b)
V1A [V] I1A [A] P1A [W] V2M [V] V1M [V] I1M [A] P1M [W] V2A [V]
10.18 0.2655 1.1 2.386 10.16 0.511 1.271 5.95
20.33 0.4835 3.252 6.31 20.91 1.014 4.208 17.41
30.18 0.6845 5.781 11.83 32.04 1.518 8.716 31.08
40.25 0.8989 9.135 17.69 39.14 1.813 12.01 39.9
50.31 1.114 13.209 23.61 50.73 2.374 19.61 54.3
60.22 1.3276 17.852 29.57 60.25 2.831 26.9 66.2
70.37 1.5553 23.55 35.5 70.12 3.328 36.19 78.7
80.4 1.7814 30 41.5 80.13 3.867 47.87 91.6
90.13 1.9985 36.59 47.4 90.17 4.435 61.67 104.1
100.1 2.24 45.29 53.3 100.4 5.11 79.51 116.8
110.5 2.499 55.3 59.7 110.2 5.935 101.8 128.9
120.8 2.772 66.6 65.8 120.3 7.139 137.9 140.9
130.3 3.051 79.62 71.5 131 8.945 201.9 152.4
140.6 3.386 96.06 77.8 139.9 10.89 280 160.7
150.9 3.764 114.75 84
161 4.222 141.32 89.8
r’dr L’ldr
Idm
Lmd I’drr
Rmd
Idrr- Idm
(a)
rqs Llqs r’qr L’lqr
Iqm
Iqss I’qrr
Vqss R mq Lmq
Vqss
I qss
( ) (
′ + j ωe Llqs + ωe Llqr
= Z qsc = Rqsc + jX qsc ≈ rqs + rqr ′ ) (3.115)
From the stationary test result in Table 3.1b and given rqs, rqr can be calculated using the
following procedure.
V1M
Z qsc = = Rqsc 2 + X qsc 2 (3.116)
I 1M
P1M
Rqsc = (3.117)
I 1M 2
From the data in Table 3.1a, the auxiliary winding rotor resistance, rdr, can be
found. When Iqs = 0, and ωr = 0 the equivalent circuit is shown in Figure 3.5.
Vdss
I dss
= Z dsc = Rdsc + jX dsc ≈ (rds + rdr (
′ ) + jωe Llqs + Llqr
′ ) (3.120)
From the stationary test result in Table 3.1a and given rds, rdr can be calculated using the
following procedure.
V1 A
Z dsc = = Rdsc 2 + X dsc 2 (3.121)
I1 A
260
Idm
Idss
Vdss Lmd I’drr
Rmd
Iqrr- Iqm
(b)
Figure 3.5. Equivalent circuit of single-phase induction machine with Iqs = 0, and ωr = 0
(a) d-axis, (b) q-axis.
P1 A
Rdsc = (3.122)
I1 A2
The turn ratio, Ndq, can be found be using the following relationship,
X dsc 1
N dq = with Nqd is the inverse of Ndq as N qd = .
X qsc N dq
From the data in Table 3.2b, the main winding parameter can be found.
Ndqωe λ’ qrr
r’dr L’ldr
- +
Idm
Lmd I’drr
Rmd
Idrr- Idm
(a)
r’qr Nqd ωe λ’ drr
rqs Llqs L’lqr
+
Iqm
Iqss I’qrr
Vqss R mq Lmq
2 2
N dq Rmq X mq
Z dq =
( )
(3.125)
4 R 2 + r 2 + 4r R + 4 X 2 × X 2 K
mq qr qr mq lqr mq 2
× Rmq + X mq
2
(
+ 8X X + 4X 2 + r2 × R2
lqr mq lqr qr mq )
Rmq Vq1 Vq 2
Rqss = rqs + X mq
2
+ J q1 + J q2 (3.126)
2
Rmq + 2
X mq Vq21 + Vq22 Vq21 + Vq22
2
Rmq Vq1 Vq 2
X qss = X lqs + X mq + J q2 − J q1 (3.127)
2
Rmq + 2
X mq Vq21 + Vq22 Vq21 + Vq22
J q1 = X mq
2 3
Rmq rqr (3.128)
J q 2 = X mq
3 2
Rmq rqr (3.129)
262
2 X mq
2
X lqr − X mq
2
rqr − 2 X mq
2
Rmq − 4 X mq X lqr Rmq − 2 X mq Rmq
2
K
Vq1 = r X (3.130)
− 2X r R + r R2 − 2X R2 qr mq
mq qr mq qr mq lqr mq
2 X mq
2
Rmq + X mq
2
rqr + 2 X mq
2
X lqr − 2 X mq rqr Rmq − 2 X mq Rmq
2
K
Vq 2 = r X (3.131)
+ 4X X R − r R2 − 2X R2 qr mq
mq lqr mq qr mq lqr mq
From the synchronous test result in Table 3.2b the following Zdq, Zqss, Rqss, and Xqss can
V2 Aa
Z dq = (3.132)
I1Ma
V1Ma
Z qss = (3.133)
I1Ma
P1Ma
Rqss = (3.134)
I1Ma 2
Given the constant values in Equations 3.132-3.135 the following unknowns Lmq, Rmq,
Llqr, and Llqs can be determined. The nonlinear Equations 3.125-3.131 are solved using
From the data in Table 3.2a, the auxiliary winding parameter can be found.
Ndqωe λ’ qrr
rds Llds r’dr L’ldr
- +
Idm
Idss
Vdss Lmd I’drr
Rmd
(a)
Iqrr- Iqm
(b)
Figure 3.7. Equivalent circuit of single-phase induction machine with Iqs = 0, and ωr = ωe
(a) d-axis, (b) q-axis.
N qd Rm2d X md
2
(3.136)
Z qd =
(
md dr dr md ldr md )
4 R 2 + r 2 + 4 r R + 4 X 2 * X 4 + 8R 2 X X 3 K
md ldr md
( )
( 3
(
+ 4 Rmd + 4 Rmd rdr + 8 X ldr + 2rdr * Rmd * X md
4 2 2 2 2
)K
)
+ 8Rmd
4
(
X md X ldr + rdr2 + 4 X ldr
2 4
* Rmd )
Rmd Vd 1 Vd 2
Rdss = rds + X md
2
+ J d1 + Jd2 (3.137)
2
Rmd + 2
X md Vd21 + Vd22 Vd21 + Vd22
2
Rmd Vd 1 Vd 2
X dss = X lds + X md + Jd2 − J d1 (3.138)
2
Rmd + 2
X md Vd21 + Vd22 Vd21 + Vd22
J d 1 = − rdr X md
2 3
Rmd (3.139)
J d 2 = − rdr X md
3 2
Rmd (3.140)
(
Vd 1 = − 2 X md
3
X ldr − 3rdr X md
2
Rmd − 4 X md
2 2
Rmd − 6 X ldr Rmd
2
X md + rdr Rmd
3
× rdr ) (3.141)
264
rdr X md
3
+ 2 X md
3
Rmd + 6 X ldr X md
2
Rmd − 3rdr Rmd
2
X md − 2 X md Rmd
3
K
Vd 2 = × r (3.142)
− 2 X R3 dr
ldr md
From the synchronous test result in Table 3.2a the following Zqd, Zdss, Rdss, and
V2 Ma
Z qd = (3.143)
I1 Aa
V1 Aa
Z dss = (3.144)
I1 Aa
P1 Aa
Rdss = (3.145)
I1 Aa 2
Given the constant values in Equations 3.143-3.146 the following unknown Lmd,
Rmd, Lldr, and Llds can be determined. The nonlinear Equations 3.136-3.142 are solved
The single-phase induction motor used was rated at 0.5 hp, 115 V, 7.8 A, and
1725 rpm. The q-d magnetizing inductances of the SPIM as a function of input voltage
are shown in Figure 3.8. The q-d magnetizing inductance increases steadily with increase
in input voltage then falls steadily with additional input voltage as the machine goes into
saturation.
Figure 3.9 shows the q-d core loss resistances also as a function of the input
voltage for the SPIM. It can be noticed from the graph the variation of the core loss
resistances with the input voltage. The value increases then decreases as the machine
The sum of q-d axis stator and rotor leakage inductances is shown in Figure 3.10
as a function input voltage. The value decreases with increase of terminal voltage due to
0.2
0.18
o Lmd
0.16
(H)
0.14
Lmd
0.12
Lmq
0.1
0.08
* Lmq
0.06
0.04
50 100 150 200
Voltage (V)
350
o Rmd
300
Rmd (Ohm)
250
200
Rmq
150
* Rmq
100
Figure 3.9. Experimental values of q-d core loss resistances vs peak input voltage.
0.03
o Ldsc
0.025
Ldsc (H)
0.02
Lqsc
0.015
* Lqsc
0.01
Figure 3.10. Experimental values sum of q -d axis leakage inductances vs peak input
voltage.
267
CHAPTER 4
4.1 Introduction
In this chapter, the steady-state analysis and simulation of the battery inverter
The battery inverter single-phase induction generator system with impedance load
will find applications in isolated areas. The generator system can be used as a source of
This chapter gives a description of the generator system, dynamic and steady-state
The first part of this chapter deals with generator system description and the
development of the mathematical model for the generator system feeding an impedance
Vd Single
+ PWM phase
Battery Cd SPIG Impedance
Inverter
- Load
70
268
load using q-d stationary reference frame transformation. Next, the dynamic
mathematical model developed is used for the simulation of the generator system with an
impedance load. The simulation results are then discussed with the experimental results.
is then used to predict the steady-state variables of the generator system with an
impedance load. The comparison between experimental results and predicted results of
the generator system with impedance load will be examined. The measured state
generator. The generator system schematic is shown in Figure 4.2. The bipolar voltage
switching scheme is used in the switching of the bridge IGBT transistors. The full-bridge
bipolar PWM inverter is fed with a battery connected through a capacitor Cd to the
excitation for the single-phase induction generator, the inverter also supplies reactive
power to the load [13, 16]. With the single-phase induction generator excited, it supplies
real power to the load. The battery acts as a source or sink of real power depending on the
real power requirement of the load and generator. When the real power supplied by the
generator exceeds the load power requirement and losses, the excess power charges the
battery through the bi-directional PWM inverter. The battery supplies real power when
269
the load power demand exceeds that provided by the generator to the load. One
advantage of this system topology is that controlling the modulation index of the inverter
Load
R o Lo
TA+ TB+
Cq
Aux. Winding
Vc Cd
Main Winding
TA- TB-
Figure 4.2. Schematic diagram of the single-phase induction generator with a battery-
PWM inverter system.
The q-d equivalent circuit for the generator system is shown in Figure 4.3 with
Vbp
C bp pV bp = I s − (4.1)
rbp
V b1
C b1 pV b1 = I s − (4.2)
rb1
The output battery voltage and open circuit battery voltage are Vd and Vbpo, respectively.
The current flowing out of the battery is Is, while rbt is the equivalent resistance of the
parallel/series battery connection. The parallel circuit of Cb1 and rb1 is used to describe
the energy and voltage during charging or discharging. Cbp is connected in parallel with
rbp to simulate the self-discharging of the battery. The battery output voltage is Vd. The
derivative of the state with respect to time is denoted as p. The equations combining the
state variables of the battery, PWM inverter and the auxiliary winding are
pVd =
1
(I s − I ds S a ) (4.4)
Cd
where Sa is the inverter switching function and Cd is the input filtering capacitance of the
inverter. The auxiliary windings input voltage and current are Vds and Ids, respectively.
The q-d equations of the single-phase induction generator and the impedance load in the
pλ ds = V ds − rds I ds (4.6)
271
pλ qs = V qs − rqs I qs (4.7)
pVqs = −
1
Cq
(
I qs + I qo ) (4.10)
pI qo =
1
Lo
(
V qs − Ro I qo . ) (4.11)
The main winding voltage is Vqs, while the q-axis and load currents are Iqs and Iqo,
respectively. The ratio of the number of turns of the q-axis winding and the d-axis
winding is denoted as Nqd. The inverse of is Nqd is denoted as Ndq. The stator q and d axes
currents are Iqs and Ids, respectively, while those of the rotor referred circuits are I'qr and
I'dr, respectively. The stator q and d axes flux linkages are λqs and λds, respectively, while
those of the rotor referred circuits are λ'qr and λ'dr, respectively. The q-d flux linkages are
′
λ qs = L qs I qs + Lmq I qr (4.12)
′
λ ds = Lds I ds + Lmd I dr (4.13)
λ ′qr = L qr
′ I qr
′ + Lmq I qs (4.14)
λ ′dr = Ldr
′ I dr
′ + Lmd I ds (4.15)
′ + Lmd . The
′ + Lmq , and Ldr = Lldr
where Lqs = Llqs + Lmq , Lds = Llds + Lmd , Lqr = Llqr
referred rotor q-d leakage inductance are L'lqr and L'ldr, respectively. The referred rotor q-
d resistance are r'qr and r'dr, respectively. The referred stator q-d leakage inductances are
L'lqs and L'lds, respectively, with the q-d magnetizing inductances given, respectively, as
272
Lmq and Lmd. The stator per-phase resistance for the q-winding and the corresponding
value for the d-axis winding are rqs and rds, respectively. The load resistor and inductor
are denoted as Ro and Lo, respectively. The electrical angular speed is ωr.
2
(
Te = P N dq λ ′qr I dr )
′ − N qd λ ′dr I qr
′ . (4.17)
The moment of inertia and the number of poles of the generator are J and P, respectively,
while the driving and generated electromagnetic torques are To and Te, respectively.
waveforms for battery inverter single-phase induction generator feeding a resistive load.
MATLAB/Simulink was used in the simulation of the dynamic Equations in 4.1 to 4.15.
Two different simulation and experimental conditions are examined. The first condition
considers the generator system in the linear modulation region ( 0 ≤ M a ≤ 1 ) while the
This section examines the simulation and experimental waveforms of the battery
inverter generator system feeding a resistive load in the linear modulation region. The
273
generator system parameters are shown in appendix A. The experiment and simulation
were carried out with a generator rotor speed of 1825 rpm and a load of resistance of 40.9
ohms. The modulation index was set at a value of 0.75 with a carrier frequency of 2 kHz
and modulating frequency of 60 Hz. The battery voltage and load capacitor is 144 V and
Figure 4.4 shows the simulated and measured steady-state waveforms for the
load voltage. A comparison the simulated and measured steady-state load voltage
waveforms shows the load frequency and magnitude are quite close. The results indicate
that the model used in the simulation predicted fairly well the laboratory experimental
setup.
The steady-state waveform for the generator main winding current is shown in
Figure 4.5. As we can observe from Figure 4.5, the main winding current is free of PWM
inverter induced harmonics, which makes this scheme good, and does not need additional
filtering. The result also indicates that the experiment differs a little in magnitude from
100 100
Main Winding Voltage (V)
50 50
Load Voltage (V)
0 0
-50 -50
-100 -100
0.945 0.95 0.955 0.96 0.965 0.97 0.975 0.02 0.025 0.03 0.035 0.04 0.045 0.05
Time (s) Time (s)
I. Simulation II. Experiment
Figure 4.4. Load voltage steady-state waveforms. Modulation index = 0.75, load
impedance = 40.9 ohms, rotor speed = 1825 rpm
274
6 6
4 4
0 0
-2 -2
-4 -4
-6 -6
0.95 0.955 0.96 0.965 0.97 0.975 0.98 0.015 0.02 0.025 0.03 0.035 0.04 0.045
Time (s) Time (s)
I. Simulation II. Experiment
Figure 4.5. Main winding current steady-state waveforms. Modulation index = 0.75, load
impedance = 40.9 ohms, rotor speed = 1825 rpm.
200 200
150 150
Auxiliary Winding Voltage (V)
100 100
50 50
0 0
-50 -50
-100 -100
-150 -150
-200 -200
0.95 0.96 0.97 0.98 0.99 1 0.01 0.02 0.03 0.04 0.05
Time (s) Time (s)
I. Simulation II. Experiment
Figure 4.6. Auxiliary winding voltage steady-state waveforms. Modulation index = 0.75,
load impedance = 40.9 ohms, rotor speed = 1825 rpm.
the single-phase induction generator are shown in Figure 4.6. The experimental result
shown Figure 4.7. As expected, this current is not free from the PWM inverter induced
harmonic.
275
The simulated and experimental current flowing through the load are shown in
Figure 4.8; as in Figure 4.5 it is free from the harmonic produced by the PWM inverter. It
can also be seen in Figures 4.4 and 4.6 that Figure 4.8 has the same frequency of the
modulating signal of the inverter (60 Hz). This result clearly shows that the load
1.5
0.5
-0.5
-1
-1.5
-2
0 0.01 0.02 0.03 0.04 0.05 0.06
Time (s)
I. Simulation II. Experiment
Figure 4.7. Auxiliary winding current steady-state waveforms. Modulation index = 0.75,
load impedance = 40.9 ohms, rotor speed = 1825 rpm.
2 2
1.5 1.5
1 1
Load Current (A)
Load Current (A)
0.5 0.5
0 0
-0.5 -0.5
-1 -1
-1.5 -1.5
-2 -2
0.96 0.965 0.97 0.975 0.98 0.985 0.99 0.01 0.015 0.02 0.025 0.03 0.035 0.04
Time (s) Time (s)
I. Simulation II. Experiment
Figure 4.8. Load current steady-state waveforms. Modulation index = 0.75, load
impedance = 40.9 ohms, rotor speed = 1825 rpm.
276
This section examines the simulation and experimental waveform of the battery
inverter generator system feeding a resistive load in the overmodulation region. The
overmodulation region is when modulation index (Ma) is greater than one (Ma>1). The
generator system parameters are shown in appendix A. The experiment and simulation
were carried out with modulation index of 1.13, a load resistance of 41.2 ohms, and a
generator rotor speed of 1830 rpm. The triangular wave signal frequency (carrier
frequency) was set at 2 kHz and sinusoidal wave signal frequency (modulating
frequency) was set at 60 Hz. The battery voltage and load capacitor are 144 V and 180
µF, respectively.
Experimental and simulation results of the load voltage under the overmodulation
range are shown in Figure 4.9. The simulation result compares favorably with the
150
100
Main Winding Voltage (V)
50
-50
-100
-150
0.005 0.01 0.015 0.02 0.025 0.03 0.035
Time (s)
I. Simulation II. Experiment
Figure 4.9. Load voltage steady-state waveforms. Modulation index =1.13, load
impedance = 40.9 ohms, rotor speed = 1830 rpm.
277
10
-5
-10
0.015 0.02 0.025 0.03 0.035 0.04 0.045
Time (s)
I. Simulation II. Experiment
Figure 4.10. Main winding current steady-state waveforms. Modulation index =1.13, load
impedance = 40.9 ohms, rotor speed = 1830 rpm.
200
150
Auxiliary Winding Voltage (V)
100
50
-50
-100
-150
-200
0.01 0.015 0.02 0.025 0.03 0.035
Time (s)
I. Simulation II. Experiment
Figure 4.11. Auxiliary winding voltage steady-state waveforms. Modulation index =1.13,
load impedance = 40.9 ohms, rotor speed = 1830 rpm.
The simulated and experimental main winding currents are shown in Figure 4.10.
As expected, the inverter induced harmonic are not present. The simulated and
experimental results are quiet close with a frequency of 60 Hz, the modulating signal
frequency.
278
-1
-2
-3
0.015 0.02 0.025 0.03 0.035 0.04 0.045
Time (s)
I. Simulation II. Experiment
Figure 4.12. Auxiliary winding current steady-state waveforms. Modulation index =1.13,
load impedance = 40.9 ohms, rotor speed = 1830 rpm
The simulated and experimental auxiliary winding voltages are shown in Figure
4.11. The results indicate clearly that the PWM inverter is operating in the
overmodulation range. The measured waveform of the auxiliary winding voltage peak is
Figure 4.12 illustrates the simulated and experimental auxiliary winding current
waveforms in the overmodulation range. The PWM inverter switching can be observed
The generator system load current is shown in Figure 4.13 for overmodulation
range. The simulated and experimental load currents are seen to be free from the PWM
inverter induced harmonics and its frequency 60 Hz the same as the modulating signal
frequency. It also indicated that the simulated waveform compares well with the
experimental waveform.
279
-1
-2
-3
-4
0.01 0.015 0.02 0.025 0.03 0.035 0.04
Time (s)
I. Simulation II. Experiment
Figure 4.13. Load current steady-state waveforms. Modulation index =1.13, load
impedance = 40.9 ohms, rotor speed = 1830 rpm.
equations were derived by using harmonic balance technique. In order to reduce the
complexity of the derivation of the generator system steady-state equation the following
(ii) The battery current, Is and inverter input voltage Vd, have second
(
λ qs = Re λ qsse jωet ) (4.18)
280
(
λ ds = Re λ dsse jωet ) (4.19)
(
λ′qr = Re λ′qrr e jωet ) (4.20)
(
λ′dr = Re λ′drre jωet ) (4.21)
(
Vqs = Re Vqsse jωet ) (4.22)
(
Vds = Re Vdsse jωet ) (4.23)
(
I qo = Re I qqoe jωet ) (4.24)
(
Sa = Re Me jωet ) (4.25)
(
I ds = Re I dsse jωet ) (4.26)
(
Vd = Re Vdo + Vd1e j 2ωet ) (4.27)
(
I s = Re I so + I s1e j 2ωet ) (4.28)
(
I qs = Re I qsse jωet ) (4.29)
(
′ = Re I qrre jωet
I qr ) (4.30)
(
′ = Re I drre jωet
I dr ) (4.31)
where Vdss, Vqss, λdss, λqss, λ'qrr, λ'drr, Iqqo, Iqss, Idss, I'qrr, and I'drr are complex peak
quantities.
[ ] [ ] [
Vds = Re[Vdo ] ∗ Re Me jωet + Re Me jωet ∗ Re Vd1e j 2ωet ;] (4.32)
281
using the identity in Equation 3.88 and comparing terms Equation 4.32 becomes
[ ] [
Re Vdsse jωe t = Re MVdoe jωe t + ] 12 ( Re[M V
*
d 1e
jω e t
]+ Re[MV
d 1e
j 3ωe t
]) . (4.33)
Since the third harmonic content is assumed to be small it can be ignored resulting in
MV d1
Vdss = MV do + . (4.34)
2
Note that the magnitude of the left-hand side is equal to the right-hand side.
[ ] [
pVd = Re[pVdo ] + Re pVd 1e j 2ωet + Re j 2ωeVd1e j 2ωet . ] (4.35)
pVd =
1
Cd
Re[I so ] + Re[I s1e j 2ωet
] { [
−
1
2
] [
Re M * I dss + Re MI dss e j 2ωet .
]} (4.36)
1 Re[MI dss ]
pVdo = I so − (4.37)
Cd 2
1 MI dss
pVd1 + j 2ωeVd 1 = I s1 − 2 . (4.38)
Cd
pVqss = −
1
Cq
( )
I qss + I qqo − jωeVqss (4.43)
pI qqo =
1
Lo
( )
V qss − R o I qqo − jω e I qqo . (4.44)
Under steady-state operating conditions, the state variables are constant, making the
derivatives in Equations 4.35- 4.44 above to be equal to zero. Then Equations 4.37 and
4.38 becomes
MI
I so = Re dss (4.45)
2
MI dss
I s1 = + j 2ω e C d V d 1 . (4.46)
2
Vd = I s R d (4.47)
Vdo = I so Rd (4.48)
Vd1 = I s1 R d . (4.49)
MR d I dss
Vd 1 = . (4.50)
2(1 − j 2ω e C d Rd )
M 2 Rd I dss
Vdss = MVdo + . (4.51)
4(1 − j 2ω e C d Rd )
M 2 Rd I dss
MVdo + − rds I dss − jω e (Lds I dss + Lmd I drr
′ )= 0 . (4.52)
4(1 − j 2ω e C d Rd )
( ′ =0
Vqss − rqs I qss − jωe Lqs I qss + Lmq I qrr ) (4.53)
′ I qrr
− N dq ω r Lqr (
′ + Lmq I qss − rdr
′ I drr )
′ − jωe (Ldr ′ + Lmd I dss ) = 0
′ I drr (4.55)
1
Cq
( )
I qss + I qqo + jω eVqss = 0 (4.56)
1
Lo
( )
Vqss − Ro I qqo − jωe I qqo = 0 . (4.57)
relating the magnetizing airgap flux to the magnetizing inductance Lmq and Lmd. For the
1 h.p. generator, Lmq and Lmd are related to the airgap flux linkages by [10]
[
λ m = λ2qm + λ2dm ]
1/ 2
(4.58)
1
Lmq
(
− 15.177 − 17.62 λ m + 22.67λ2m = 0 ) (4.59)
1
Lmd
(
− 6.29 − 3.28 λ m + 3.042λ2m = 0 . ) (4.60)
4
(
Te1 = P N dqλ′qrr I drr
′* − N qd λ′drr I qrr
′* . ) (4.61)
The nonlinear Equations 4.52-4.61 are solved for the state variables using MATLAB.
284
The constant parameters for the generator system are shown in appendix A. The
experiment was carried out with generator rotor speed of 1830 rpm and load resistances
of 10.2 and 55.6 ohms. The battery voltage and load capacitor are 144 V and 210 µF,
respectively.
Figure 4.14 shows how measured and calculated main winding voltages of the
single-phase induction generator vary as a function of the modulation index. The result
shows that in the linear region (Ma<1) the main winding voltage increases linearly by but
in the over modulation region the main winding voltage becomes fairly constant. Hence
in applications that require higher load voltages such as heating, the modulation index
could be increased appropriately to give a higher voltage which leads to more energy.
140
120
Main Winding Voltage (V)
100
80
60
40
10.2 Ohms
20 55.6 Ohms
Calc
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.14: Measured and calculated main winding voltage as a function of modulation
index. Constant rotor speed = 1830 rpm.
285
12
10
2 10.2 Ohms
55.6 Ohms
Calc
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.15: Measured and calculated main winding current as a function of modulation
index. Constant rotor speed = 1830 rpm.
Figure 4.15 shows the variation of the measured and calculated main winding
The current increases linearly in the linear region and becomes relatively constant then in
The measure and calculated rms load currents as a function of the modulation are
shown in Figure 4.16. The result indicates that the steady state model predicts the load
current more accurately at higher resistor value. This is attributed to the fact an inverter
loss, which is not accounted for in the model, increases with increasing load/inverter
output current.
286
7 10.2 Ohms
55.6 Ohms
Calc
6
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.16: Measured and calculated load current as a function of modulation index.
Constant rotor speed = 1830 rpm.
600
10.2 Ohms
500 55.6 Ohms
Calc
Output Power (W)
400
300
200
100
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.17: Measured and calculated load power as a function of modulation index.
Constant rotor speed = 1830 rpm.
Figure 4.17 shows the variation of the measured and calculated load powers as a
function of the modulation index. The result indicates that the steady state model predicts
the load power more accurately at higher resistor value. This is attributed to the fact that
287
inverter loss, which is not accounted for in the model, increases with increasing load
current.
Figures 4.18 and 4.19 show the variation of the measured and calculated auxiliary
winding voltages and currents as a function of the modulation index, respectively. The
calculated value at 10.2 ohms is relatively different from the measurement due to inverter
switching and inverter losses that are not taken into account in the model. This difference
is higher at a small value of load impedance (large load current). The result indicates that
the steady state model predicts the load power more accurately at higher resistor value.
This is attributed to the fact that losses are small at lower values of load currents.
200
Auxiliary Winding Voltage (V)
150
100
50
10.2 Ohms
55.6 Ohms
Calc
0
0 0.5 1 1.5 2 2.5
Modulation Index
1.5
10.2 Ohms
55.6 Ohms
1 Calc
0.5
0
0 0.5 1 1.5 2 2.5
Modulation Index
8
10.2 Ohms
6 55.6 Ohms
Calc
Battery Current (A)
-2
-4
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.20: Measured and calculated battery current as a function of modulation index.
Constant rotor speed = 1830 rpm.
index are shown in Figure 4.20. The result indicates that the steady state model predicts
the load power more accurately at higher resistor value. This is attributed to the fact that
289
losses are small at lower value of load current. The negative battery current (55.6 Ω)
indicates the condition during which the generator system absorbs real power while the
positive battery current (10.2 Ω) shows that the generator system supplies real power to
the load.
This section discusses the measured steady state performance of the battery
inverter generator system feeding a resistive load. The experiment setup is carried under
two different load resistances of 41.2 and 20.8 ohms. The generator rotor speed and
triangular wave signal frequency was kept at 1840 rpm and 2 kHz, respectively. The
sinusoidal wave signal frequency was set at 60 Hz. The battery voltage and load capacitor
index is shown in Figure 4.21. The load voltage increases with increase in the load
resistance. The system increases linearly in the linear region then become non-linear in
Figure 4.22 shows the single-phase induction generator main winding current as a
function of the modulation index. The current gradually increases in the linear region
120
100
60
40
20
20.8 Ohms
41.2 Ohms
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.21: Measured load voltage as a function of modulation index. Constant rotor
speed = 1840 rpm.
7
Main Winding Current (A)
1 20.8 Ohms
41.2 Ohms
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.22: Measured main winding current as a function of modulation index. Constant
rotor speed = 1840 rpm.
291
The output power from the generator system is shown in Figure 4.23 as a
variation of modulation index. The power increases linearly for each resistance value in
the linear region and it becomes fairly constant in the over modulation region.
The generator system measured steady-state characteristic for the load current as
a function of modulation index for different load values, as shown in Figure 4.24. The
load current is relatively constant in the over modulation region while it increases linearly
winding voltage as a variation of the modulation index. The load resistor value of 41.2
ohms has more voltage in the auxiliary winding voltage compared with the load resistor
value of 20.8 ohms. The voltage across the auxiliary winding is fairly constant in the over
300
250
200
Output Power (W)
150
100
50
0 20.8 Ohms
41.2 Ohms
-50
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.23: Measured Output Power as a function of modulation index. Constant rotor
speed = 1840 rpm.
292
2.5
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.24: Measured Load Current as a function of modulation index. Constant rotor
speed = 1840 rpm.
160
140
Auxiliary winding Voltage (V)
120
80
60
40
0 0.5 1 1.5 2 2.5
Modulation Index
function of the modulation index is shown in Figure 4.26. The current flow through the
293
auxiliary winding is increasing over the range of modulation index. This is due to the
presence of higher harmonic in the auxiliary winding current as shown in the dynamic
The generator system battery current for both resistor values as a function of
modulation index in shown in Figure 4.27. At load resistance of 41.2 ohms the battery
current is negative. This is due to the fact that the battery absorbs the excess power
produced by the single-phase induction generator not required by the load. On the other
hand, when the load resistance is 20.8 ohms, the battery current is positive. Under this
condition the single-phase induction generator gives reduced power required by the load,
hence the battery augments the power of the induction generator by providing extra
power.
5
20.8 Ohms
4.5 41.2 Ohms
Auxiliary Winding Current (A)
3.5
2.5
1.5
1
0 0.5 1 1.5 2 2.5
Modulation Index
-1
-2
-3
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.27: Measured battery current as a function of modulation index. Constant rotor
speed = 1840 rpm.
This section discusses the measured steady state performance of the battery
inverter generator system feeding a resistive load under variable generator rotor speed.
The experiment is carried out with load resistance of 20.8 ohms and generator rotor speed
of 1840 and 1850 rpm; the triangular wave signal frequency (carrier frequency) was set at
2 kHz and sinusoidal wave signal frequency (modulating frequency) was set at 60 Hz.
The battery voltage and load capacitor are 96 V and 180 µF, respectively. The parameters
The generator system load voltage as a function of the modulation index is shown
in Figure 4.28. The graph shows that with 10 rpm increase in the single-phase induction
generator speed the load voltage is almost doubled. Also the load voltage is relatively
295
constant in the overmodulation region while in the linear region the load voltage
increases linearly with the modulation index. Hence, the generator system load voltage
can be increased either by increasing the modulation index or increasing the induction
Figure 4.29 shows the variation of the main winding current of the induction
generator as a function of the modulation index. The main winding current at 1850 rpm is
about twice the induction generator rotor speed at 1840 rpm. The main winding current
increases linearly in the linear region then gradually becomes constant in the over
modulation region.
shown in Figure 4.30. It is observed that at higher rotor speed there is more output power.
In addition, more output power could be obtained by increasing the modulation index.
100
80
Load Voltage (V)
60
40
20 1840 rpm
1850 rpm
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.28: Measured load voltage as a function of modulation index. Constant load
impedance = 20.8 ohms.
296
2 1840 rpm
1850 rpm
1
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.29: Measured main winding current as a function of modulation index. Constant
load impedance = 20.8 ohms.
The generator system load current as a function of the modulation index is shown
in Figure 4.31. The load current is higher for generator rotor speed of 1850 rpm as
compared to 1840 rpm (Figure 4.31). Also, a higher load current can be achieved in the
Figure 4.32 shows the variation of the induction generator measured auxiliary
winding voltage as a function of modulation index. The voltage across the auxiliary
winding voltage is increased with increase in the rotor speed of the generator. A higher
350
300
250
150
100
50
1840 rpm
0 1850 rpm
-50
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.30: Measured Output Power as a function of modulation index. Constant load
impedance = 20.8 ohms.
4
Load Current (A)
1 1840 rpm
1850 rpm
0
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.31: Measured Load Current as a function of modulation index. Constant load
impedance = 20.8 ohms.
298
140
1840 rpm
1850 rpm
100
80
60
40
0 0.5 1 1.5 2 2.5
Modulation Index
the modulation index is shown in Figure 4.33. Due to the presence of higher harmonic in
the auxiliary winding current, as shown in the dynamic waveforms in Figures 4.7 and
4.12, the current increases relatively over the range of modulation index.
modulation index. It is observed that the figure shows the two modes of operation of the
generator system (when the battery absorbs power and when the battery provides power).
At generator rotor speed of 1850 rpm battery current is negative. This is due to the fact
that the battery absorbs the excess power not required by the load produced by the single-
phase induction generator. On the other hand, when the generator rotor speed is 1840
rpm, the battery current is positive. Under this condition the single-phase induction
299
generator gives lower power required by the load hence the battery augments the power
4.5
1840 rpm
4 1850 rpm
Auxiliary Winding Current (A)
3.5
2.5
1.5
1
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.33: Measured auxiliary winding current as a function of modulation index.
Constant load impedance = 20.8 ohms.
1.5
Battery Current (A)
0.5
1840 rpm
1850 rpm
0
-0.5
-1
0 0.5 1 1.5 2 2.5
Modulation Index
Figure 4.34: Measured battery current as a function of modulation index. Constant load
impedance = 20.8 ohms.
300
This section discusses the measured steady state performance of the battery
inverter generator system feeding a resistive load. The modulation index of the PWM
carried out with load resistance of 41.1 ohms; the triangular wave signal frequency
(carrier frequency) was set at 2 kHz and sinusoidal wave signal frequency (modulating
frequency) was set at 60 Hz. The battery voltage and load capacitor are 144 V and 180
µF, respectively. The parameters of the test generator are given in the appendix A.
Figure 4.35 shows the auxiliary winding voltage of the induction generator as a
function of generator slip. The figure indicates that maximum auxiliary winding voltage
is obtained at about -0.05 slip under the regulated voltages considered. With a higher
160
150
Auxiliary Winding Voltage (V)
140
40 V
130 60 V
80 V
120
110
100
90
80
-0.1 -0.08 -0.06 -0.04 -0.02 0
Slip
Figure 4.35: Measured auxiliary winding voltage as a function of slip. Constant load
impedance = 41.1 ohms.
301
10
40 V
60 V
0
-0.1 -0.08 -0.06 -0.04 -0.02 0
Slip
Figure 4.36: Measured auxiliary winding current as a function of slip. Constant load
impedance = 41.1 ohms.
shown in Figure 4.36. The auxiliary winding current of the generator increases at
decreasing values of generator slip. The graph also indicates that at higher values of
induction generator slip the auxiliary winding currents at the given regulated voltage are
Figure 4.37 shows the variation of the modulation index as a function of the
generator slip. As the generator slip increases the modulation index also increases. The
graph also indicates a higher regulated output voltage can be obtained by increasing the
modulation index.
Figure 4.38 shows the measured battery current as a function of generator slip.
The generator system takes power from the battery for a range of the rotor speeds,
1.8
1.6 40 V
60 V
80 V
1.4
Modulation Index
1.2
0.8
0.6
0.4
0.2
-0.1 -0.08 -0.06 -0.04 -0.02 0
Slip
Figure 4.37: Measured modulation index as a function of slip. Constant load impedance =
41.1 ohms.
5
Battery Current (A)
40 V
60 V
80 V
-5
-0.1 -0.08 -0.06 -0.04 -0.02 0
Slip
Figure 4.38: Measured battery current as a function of slip. Constant load impedance =
41.1 ohms.
303
parameter on the generator system performance. In the study that follows, two different
PWM inverter switching scheme are considered - the unipolar voltage switching and
A proper selection of the compensating capacitor, Cq, and the generator rotor
speed can ensure that the generator system with impedance load operates in optimum
condition. It can be seen from the steady state results in section 4.4, that the generator
system has to operate in the overmodulation range to obtain higher a load voltage. The
selection of the load capacitor, Cq and the generator rotor speed should be carefully done
to ensure overall generator system efficiency and maximum power. This will be useful in
applications such as heating where load voltage frequency is not important consideration.
The generator system steady state was studied by keeping the load impedance and the
In Figure 4.39 - 4.46 the modulation index is kept constant at a value of one, the
load impedance of 50 ohms, and load frequency of 60Hz. Method A shows the PWM
with unipolar voltage switching and Method B shows the PWM with bipolar voltage
switching.
Figure 4.39 shows the contour plots of the load voltage with variation of the
generator rotor speed and load capacitor, Cq with the two PWM methods. The graph
shows that generally method B gives a higher input voltage than method A under the
304
same condition. From the graph, if maximum load voltage is desired, a higher value of Cq
has to be selected and a generator rotor speed of about 1.08 p.u. is required.
A family of main winding currents as a variation of the generator rotor speed and
load capacitor, Cq with the two PWM methods is shown in Figure 4.40. The graph shows
that as the load capacitor increases the main winding current also increases. The
350 350
119 136
300 300
100 114
250 250
Cq (uF)
Cq (uF)
200 200
80.8 91.8
48.2
150 150
42.4
61.6 70 70
100 100
61.6
Method A Method B
Figure 4.39: Contour plot of load voltage, [V] as a variation of generator rotor speed (per
unit) and load capacitor Cq.
350 350
15.5 17.6
9.59 10.9
250 250
6.63 7.53
Cq (uF)
Cq (uF)
200 200
3.66
150 150
4.16
100 100
350 350
2.38 2.71
300 300
2 2.27
250 250
Cq (uF)
Cq (uF)
200 200
1.62 0.963 1.84
0.847
150 150
1.23 1.4
1.23 1.4
100 100
350 350
309 399
300 300
234
303
250 250
160
Cq (uF)
Cq (uF)
206
200 200
85.3 85.3 110
150 150
110
100 100
maximum main winding current is obtained at a generator rotor speed of about 1.08p.u.
Figure 4.41 gives the load current contour plots as a variation of the generator
speed and load capacitor, Cq with the two PWM methods. Both methods A and B indicate
that the load current decreases as the load capacitor decreases. The maximum load
350 350
120
300 300
106 121
106 120
250 250
106
Cq (uF)
Cq (uF)
200 200
106 106 120 120
150 150
106 120
100 100
Figure 4.42 shows the load power contour as variation of the generator speed and
load capacitor, Cq with the two PWM methods. The graph clearly indicates that for
maximum power application of the generator system, the generator system must be
operated at higher value of load capacitor, at about 1.075 p.u generator rotor speed. This
The family of auxiliary winding voltage and current as variation of the generator
speed and load capacitor, Cq with the two PWM methods are shown in Figures 4.43 and
4.44. The auxiliary winding voltage is generally constant for both PWM methods. The
auxiliary winding current, on the other hand, is fairly constant as the load capacitor
changes for constant generator rotor speed. The minimum auxiliary winding current is
350 350
Cq (uF)
Cq (uF)
5.67
200 200
14.9
5.67
150 150
9.92 16.9
100 100
Method A Method B
Figure 4.44: Contour plot of auxiliary winding current, [A] as a variation of generator
rotor speed (per unit) and load capacitor Cq.
350 350
36.3 36.3
36.3
300 300 53.8
18.9 53.8 36.3
18.9 18.9
250 250
18.9
Cq (uF)
Cq (uF)
200 200
150 150
71.2 71.2
100 100
Figure 4.45 shows the generator system efficiency contour as a function of the
generator speed and load capacitor, Cq with the two PWM methods. The graph shows that
generator system efficiency is better at lower values of load capacitor and at a generator
The family of battery current curves as variation of the generator speed and load
capacitor, Cq with the two PWM methods is shown in Figure 4.46. Generally the plots
indicate that when the generator rotor speed is greater than 1.0 p.u. the battery current is
negative (absorbs real power from the load) and when it is lesser than 1.0 p.u. the battery
current is positive (supplies real power to the load) under different values of load
capacitor.
350 350
Cq (uF)
Cq (uF)
200 200
-4.66
-4.66 -6.02 -6.02
150 150
-0.576
-0.745
100 100
350 350
113 86.4
300 300
46.5
95.1
250 250
73.1
Cq (uF)
Cq (uF)
200 200
77
150 150
40.7 58.8 33.2 59.8 59.8
100 100
58.8
In Figures 4.47 - 4.49 the modulation index is kept constant at a value of one and
load frequency of 60Hz with the PWM with unipolar voltage switching (Method B) and
Figure 4.47 shows the contour plots of the load voltage with variation of the
generator rotor speed and load capacitor, Cq with the two load impedance. The graphs
generally have similar contours but differ simply in the load voltage magnitudes. As the
load impedance increases the load voltage increases. In addition, to obtain maximum load
voltage (also in Figure 4.39 method A) a higher value of Cq has to be selected and a
Figure 4.48 shows the contour plots of the generator system efficiency as a
function of the generator rotor speed and load capacitor, Cq with two load impedance.
Examining Figure 4.45 (Method A) and Figure 4.48; the generator system efficiency is
seen to be better at lower value of load impedance at fixed value of load capacitor and
generator rotor speed. The graphs also indicate that the generator system efficiency is
higher at about (0.99 to 1.01 p.u.) generator rotor speed for different values of load
capacitor.
350 350
300 55 300
55 70
37.6
70
250 250
19.9 37.5
Cq (uF)
Cq (uF)
200 200
72.6 53.8
150 150
37.5 53.8
100 100
19.9
350 350
300 300
-8.26 -5.9
250 250
3.61 -0.346 4.11
Cq (uF)
Cq (uF)
-4.3
200 200
-4.3 -2.56
150 150
0.775
Figure 4.49 shows the battery current contour curves as variation of the generator
rotor speed and load capacitor, Cq under different load condition. Considering Figure
4.46 (Method A) with Figure 4.49, the battery current is observed to be more positive (or
higher) as the load impedance decreases at constant load capacitor and generator rotor
speed. When the battery current is negative it indicates that real power is being absorbed
by the battery and when the battery current is positive it shows that the battery supplies
4.6 Transient and Dynamic Performance for the System Feeding an Impedance
Load
This section examines the transient and dynamic performance of the battery
inverter single-phase induction generator feeding a resistive load. The transient and
dynamic studied gives a better understanding of the operation of the generator system and
also shows how the generator system can perform with either variation in load or speed.
311
The generator system start-up process is simulated and the response of the generator
system to changes in load is examined. Next the system dynamic performance with the
This section examines the start-up process for the battery inverter generator
system feeding a resistive load. The simulation was carried out with generator rotor speed
of 1860 rpm and a load of resistance of 20 ohms. The modulation index was set a value
The system load voltage build-up is shown in Figure 4.50. The load voltage
waveform increases quickly until it reaches a steady-state peak value of about 80 volts.
Figure 4.51 shows the start-up process of the system main winding current. The main
The auxiliary winding current waveform build-up is shown in Figure 4.52. The
current decreases fast to a steady state peak value of about 4.0A. The transient waveform
of the battery current is shown in Figure 4.53 during start up. The battery current is
initially high then decreases to a steady state value. Initially the battery current is positive
indicating that the battery supplies real power to meet the load requirement, then at
steady state the battery current is negative indicating that the battery is absorbing real
power.
312
100
50
-50
-100
0 0.05 0.1 0.15 0.2
Time (s)
Figure 4.50 Load voltage start-up waveform. Modulation index = 0.8, load impedance =
20 ohms, rotor speed = 1860 rpm.
6
Main Winding Current (A)
-2
-4
-6
-8
0 0.05 0.1 0.15 0.2
Time (s)
Figure 4.51 Main winding current start-up waveform. Modulation index = 0.8, load
impedance = 20 ohms, rotor speed = 1860 rpm.
313
15
10
-5
-10
-15
-20
0 0.05 0.1 0.15 0.2
Time (s)
Figure 4.52 Auxiliary winding current start-up waveform. Modulation index = 0.8, load
impedance = 20 ohms, rotor speed = 1860 rpm.
3
Battery Current (A)
-1
-2
-3
-4
0 0.05 0.1 0.15 0.2
Time (s)
Figure 4.53 Battery current start-up waveform. Modulation index = 0.8, load impedance
= 20 ohms, rotor speed = 1860 rpm.
314
Figure 4.54 Generator torque start-up waveform. Modulation index = 0.8, load
impedance = 20 ohms, rotor speed = 1860 rpm.
The graph shows how the generator torque falls quickly to a steady state value.
This section examines the dynamics for the battery inverter generator system
feeding a resistive load for changes in the load impedance. The simulation was carried
out with generator rotor speed of 1840 rpm, the modulation index of 0.80, a carrier
The load impedance is initial is set at a value of 30 ohms and when the system
reaches its steady state operating condition it is changed from 30 ohms to 50 ohms and
4.55.
315
The system load voltage waveform is shown in Figure 4.56. It can be seen that the
generator responded quickly to the change in the load impedance to operate in steady-
state condition. Figure 4.57 shows the main winding current waveform. The main
winding current increases as the load impedance is changed from 50 ohms to 10 ohms.
60
50
Load Resistance (Ohms)
40
30
20
10
0
0.2 0.25 0.3 0.35 0.4 0.45
Time (s)
Figure 4.55. Changes in the values of the load impedance. Modulation index = 0.8, and
rotor speed = 1840 rpm.
Figure 4.56. Load voltage waveform response to changes in load. Modulation index =
0.8, and rotor speed = 1840 rpm.
316
Figure 4.57. Main winding current waveform response to changes in load. Modulation
index = 0.8, and rotor speed = 1840 rpm.
The auxiliary winding current waveform response due to changes in load is shown
in Figure 4.58. As the load impedance is changed from 50 ohms to 10 ohms, the auxiliary
winding current decreases. The battery current waveform is shown in Figure 4.59. The
battery current is initially negative when the load impedance is at 30 and 50 ohms. The
battery current becomes positive as the battery supplies real power to augment the real
power provided by the generator to the load when the load impedance is changed from 50
ohms to 10 ohms. This clearly indicates the reliability of the system to a sudden change
The generator torque increase when the load impedance changes from a light load (50
This section examines the dynamics for the battery inverter generator system
feeding a resistive load for changes in generator rotor speed. The simulation was carried
out with a load resistance of 40 ohms, modulation index of 0.80, a carrier frequency of 2
The generator rotor speed is initially is set at a value of 1840 rpm when the
system reaches its steady state operating condition it is changed from 1840 rpm to 1900
Figure 4.62 shows load voltage waveform response to changes in the generator
rotor speed. The load voltage is maintained while the generator rotor speed is less than
the 1800 rpm (synchronous speed) as the battery supplies the needed real power required
319
by the load. Figure 4.63 shows the main winding current waveform. The main winding
current decrease as the generator rotor speed changes from 1900 rpm to 1740 rpm.
1900
Generator Rotor Speed (rpm)
1850
1800
1750
Figure 4.62. Load voltage waveform response to changes in generator rotor speed.
Modulation index = 0.8, and load resistance = 40 ohms.
320
Figure 4.63. Main winding current waveform response to changes in generator rotor
speed. Modulation index = 0.8, and load resistance = 40 ohms.
Figure 4.64. Auxiliary winding current waveform response to changes in generator rotor
speed. Modulation index = 0.8, and load resistance = 40 ohms.
rotor speed is shown in Figure 4.64. The auxiliary winding current decreases in response
321
to change in the generator rotor speed as the speed changes from 1900 rpm to 1740 rpm.
The battery current waveform response to changes in the generator rotor speed is shown
in Figure 4.65. Initially the battery absorb real power (negative current) afterwards when
the generator rotor speed changes to 1740 rpm from 1900 rpm the battery now provides
real power (positive current) to the load. This result shows an advantage of the battery-
inverter system over the conventional single-phase induction generator that will operate
as a motor if the generator rotor speed is less than the synchronous speed.
The generator torque is negative when the speed is at 1840 and 1900 rpm and becomes
Figure 4.65. Battery current waveform response to changes in generator rotor speed.
Modulation index = 0.8, and load resistance = 40 ohms.
322
Figure 4.66. Generator torque waveform response to changes in generator rotor speed.
Modulation index = 0.8, and load resistance = 40 ohms.
323
CHAPTER 5
5.1 Introduction
induction motor in appliances such as in fans, washing machines, dryers, refrigerator, etc.
worthwhile to analyze and study the performance of the battery inverter single-phase
generator system (SPIG) feeding a single-phase induction motor (SPIM) load will be
studied. This system can be applied as a source of power in a remote location where there
is no conventional utility power supply. Some of the system applications include milling,
Vd
+ PWM SPIM
Battery Cd SPIG
Inverter Load
-
126
324
The block diagram describing the battery inverter SPIG feeding SPIM is shown in
Figure 5.1.
The first part of this chapter looks into the system operation and the mathematical
model of the battery inverter system with SPIM load. The dynamic mathematical model
of the generator system developed is used in the simulation. The simulation results are
compared with the measured waveform. The steady-state equation is further developed
from the dynamic mathematical model. The resulting steady-state equation is used to
induction generator, and the load (single-phase induction motor). The schematic is shown
in Figure 5.2.
The system operation is similar to the system description in section 4.2. With the
generator connected in the fashion shown in Figure 5.2, the H bridge transistors are
switched using the bipolar voltage switching scheme. The input to PWM inverter is
obtained from a bank of battery fed through a capacitor. The capacitor, Cd ensures that
the input voltage to the inverter is kept constant. The induction generator is connected in
a manner that ensures that minimum induced inverter harmonics. The generator topology
is configured in such away that the output from the inverter is connected to the auxiliary
325
Induction Motor
aux. winding
Vds
main winding
TA+ TB+
Cq
Aux. Winding
Vc Cd Main Winding
TA-
TB
-
Induction Generator
Figure 5.2. Schematic diagram of the system with induction motor load.
winding (the d-axis) and the main winding is connected to the single-phase induction
motor load. The advantage of this scheme is that it ensures that the output voltage into the
The frequency of the input voltage of SPIM motor load is determined by the
modulating signal (sine waveform) of the PWM. Hence simply adjusting the modulating
signal appropriately the motor load changes speed. The power requirement by the motor
load is supplied from the battery or SPIG, depending on the speed of the SPIG. Also the
reactive power needed by the motor load is supplied by the inverter [13,16]. The
advantage of this isolated system is its ability to meet the motor load power demand.
When the SPIG supplies more power than required by the motor the excess power is used
326
to charge the battery and when the power provided by the generator is less than the
The equivalent circuit of the battery, PWM inverter, generator and motor is shown
in Figure 5.3. Equations 4.1 - 4.9 representing the model of the battery, inverter, and
Vbp
C bp pV bp = I s − (5.1)
rbp
V b1
C b1 pV b1 = I s − (5.2)
rb1
rbl Νdqωrλ’qr
rbs rbt rds Llds r'dr L’ldr
ω- rλ’
+ qr
- +
rbp Cbl Is Ids Lmd I’dr
Cbp Cd
Inverter Generator
Battery
(a)
Νdqmωrmλ’qrm
rdsm Lldsm r'drm L’ldrm
ωrλ’qr- +
+
- +
Vdsm Idsm Lmdm I’drm
-
Motor Load
(b)
Νqdmωrmλ’drm Ν ω λ’
L’lqrm r’qrm Llqsm rqsm rqs Llqs r’qr L’ qd r dr
lqr
+ +
Figure 5.3. The q-d equivalent circuit of the battery-PWM inverter generator system with
single-phase induction motor load. (a-b) d-axis circuit, (c) q-axis circuit.
327
pVc =
1
(I s − I ds Sa ) (5.4)
Cd
V ds = S aV c (5.5)
pλ ds = V ds − rds I ds (5.6)
pλ qs = V qs − rqs I qs (5.7)
pVqsm = −
1
Cq
(
I qs + I qsm ) (5.10)
where the motor main winding and auxiliary winding voltage are denoted as Vqsm and
Vdsm, respectively. The ratio of the number of turns of the motor q-axis winding and the
d-axis winding is denoted as Nqdm. The inverse of is Nqdm is denoted as Ndqm. The motor
stator q and d axes currents are Iqsm and Idsm, respectively, while those of the rotor
referred circuits are I'qrm and I'drm, respectively. The motor stator q and d axes flux
linkages are λqsm and λdsm, respectively, while those of the rotor referred circuits are λ'qrm
Equations 4.12 - 4.15 representing the q-d flux linkages of the generator are given as
′
λ qs = L qs I qs + Lmq I qr (5.15)
′
λ ds = Lds I ds + Lmd I dr (5.16)
λ ′qr = L qr
′ I qr
′ + Lmq I qs (5.17)
λ ′dr = Ldr
′ I dr
′ + Lmd I ds . (5.18)
′
λ qsm = Lqsm I qsm + Lmqm I qrm (5.19)
′
λ dsm = Ldsm I dsm + Lmdm I drm (5.20)
λ ′qrm = Lqrm
′ I qrm
′ + Lmqm I qsm (5.21)
λ ′drm = Ldrm
′ I drm
′ + Lmdm I dsm (5.22)
′ + Lmdm . The referred motor rotor q-d leakage inductances are L'lqrm and
Ldrm = Lldrm
L'ldrm, respectively. The referred motor rotor q-d resistance are r'qrm and r'drm,
respectively. The motor stator q-d leakage inductances are L'lqsm and L'ldsm, respectively,
with the q-d magnetizing inductances given, respectively, as Lmqm and Lmdm. The motor
stator per-phase resistance for the q-winding and the corresponding value for the d-axis
winding are rqsm and rdsm, respectively. The motor electrical angular speed is ωrm.
The dynamics of the generator turbine is given in Equations 4.16 and 4.17 and are
given as
2
(
Te = P N dq λ ′qr I dr )
′ − N qd λ ′dr I qr
′ . (5.24)
329
pωrm =
Pm
(T − Tom )
2 J m em
(5.25)
Tem =
Pm
2
(
N dqm λ ′qrm I drm ′ )
′ − N qdmλ ′drm I qrm . (5.26)
The moment of inertia and the number of poles of the generator are Jm and Pm,
respectively. The driving and generated electromagnetic torques are respected as Tom and
Tem, respectively.
waveforms of the battery inverter SPIG system feeding SPIM load. The conditions
examined include when the modulation index of the PWM inverter is in the linear
modulation region and also when the modulation index is in the overmodulation region.
MATLAB/Simulink.
5.3.1 Battery Inverter SPIG System Feeding SPIM Load (Linear Region)
This section looks at the measured and simulated waveforms when the battery
inverter SPIG is feeding a SPIM when the modulation index of PWM inverter is in the
linear region. The experiment and simulation were carried out with generator rotor speed
of 1840 rpm. The modulation index was set a value of 0.875 with a carrier frequency of
330
2 kHz and modulating frequency of 60 Hz. The battery voltage and load capacitor are 144
V and 180 µF, respectively. The parameter of the SPIM is shown in appendix B.
Figure 5.4 shows the input voltage into the single-phase induction motor under no
load consideration. Simulation and experiment waveforms are observed to have the same
characteristics.
100 100
Motor Input Voltage (V)
0 0
-50 -50
-100 -100
2.97 2.975 2.98 2.985 2.99 2.995 0.005 0.01 0.015 0.02 0.025 0.03 0.035
Time (s) Time (s)
I. Simulation II. Experiment
Figure 5.4. Induction motor input voltage steady-state waveforms. Modulation index =
0.875, rotor speed = 1840 rpm
3 3
Generator Main Winding Current (A)
2 2
1 1
0 0
-1 -1
-2 -2
-3 -3
2.96 2.97 2.98 0.025 0.03 0.035 0.04 0.045 0.05
Time (s) Time (s)
I. Simulation II. Experiment
Figure 5.5. Induction generator main winding current steady-state waveforms.
Modulation index = 0.875, rotor speed = 1840 rpm.
331
0 0
-50 -50
-100 -100
-150 -150
-200 -200
2.95 2.96 2.97 2.98 2.99 0.01 0.02 0.03 0.04 0.05
Time (s) Time (s)
I. Simulation II. Experiment
Figure 5.6. Generator auxiliary winding voltage steady-state waveforms.
Modulation index = 0.875, rotor speed = 1840 rpm.
6 6
4 4
2 2
0 0
-2 -2
-4 -4
-6 -6
2.87 2.875 2.88 2.885 2.89 2.895 2.9 0.015 0.02 0.025 0.03 0.035 0.04 0.045
Time (s) Time (s)
I. Simulation II. Experiment
Figure 5.7. Generator auxiliary winding current steady-state waveforms.
Modulation index = 0.875, rotor speed = 1840 rpm.
Figure 5.5 shows the simulated and measured steady-state generator main
voltage of the generator are shown in Figure 5.6. The experimental waveform is different
from the simulate waveform due to the harmonic content in the battery current.
332
auxiliary winding current are shown in Figure 5.7. The inverter induced harmonic is
Figure 5.8 shows the input current or the main winding current steady-state
waveform of the single-phase induction motor. The results show that the frequency of the
current is same as the modulating frequency, which in this case is 60 Hz. The results from
in Figure 5.9. The battery current waveform is negative indicating that the battery is
being charged. The experimental waveform shows the inverter induced harmonics riding
5 5
Motor Main Winding Current (A)
Motor Main Winding Current (A)
0 0
-5 -5
2.96 2.97 2.98 0.015 0.02 0.025 0.03 0.035 0.04
Time (s) Time (s)
I. Simulation II. Experiment
Figure 5.8. Induction motor main winding current steady-state waveforms.
Modulation index = 0.875, rotor speed = 1840 rpm.
333
-1.5
-1.55
-1.65
-1.7
-1.75
-1.8
5.3.2 Battery Inverter SPIG System Feeding SPIM Load (Overmodulation Region)
This section looks at the measured and simulated waveforms when the battery
inverter SPIG is feeding a SPIM when the modulation index of PWM inverter is in the
overmodulation region. The experiment and simulation were carried out with generator
rotor speed of 1840 rpm. The modulation index was set a value of 1.375 with a carrier
frequency of 2 kHz and modulating frequency of 60 Hz. The battery voltage and load
capacitor are 144 V and 180 µF, respectively. The parameter of the SPIM is shown in
appendix B.
Figure 5.10 shows the simulated and experimental motor input steady-state
voltage waveform. The waveform shows that its frequency is the same as the modulating
signal of the inverter and it is also free from the inverter induced harmonics.
main winding current is shown in Figure 5.11. The simulated and experimental
150 150
100 100
Motor Input Voltage (V)
0 0
-50 -50
-100 -100
-150 -150
2.97 2.975 2.98 2.985 2.99 2.995 0.015 0.02 0.025 0.03 0.035 0.04
Time (s) Time (s)
I. Simulation II. Experiment
Figure 5.10. Induction motor input voltage steady-state waveforms.
Modulation index = 1.375, rotor speed = 1840 rpm
4 4
Generator Main Winding Current (A)
Generator Main Winding Current (A)
3 3
2 2
1 1
0 0
-1 -1
-2 -2
-3 -3
-4 -4
2.955 2.96 2.965 2.97 2.975 2.98 2.985 0.01 0.02 0.03
Time (s) Time (s)
I. Simulation II. Experiment
Figure 5.11. Induction generator main winding current steady-state waveforms.
Modulation index = 1.375, rotor speed = 1840 rpm.
Figure 5.12 shows the simulated and experimental waveforms of the induction
generator steady-state auxiliary winding voltage. The experimental results show that the
auxiliary winding voltage is influence by the charging of the battery (that is battery is
5 5
0 0
-5 -5
-10 -10
2.955 2.96 2.965 2.97 2.975 2.98 2.985 0.005 0.01 0.015 0.02 0.025 0.03 0.035
Time (s) Time (s)
I. Simulation II. Experiment
Figure 5.13. Generator auxiliary winding current steady-state waveforms.
Modulation index = 1.375, rotor speed = 1840 rpm.
8 8
6 6
Motor Main Winding Current (A)
2 2
0 0
-2 -2
-4 -4
-6 -6
-8 -8
2.96 2.97 2.98 0.02 0.03 0.04
Time (s) Time (s)
I. Simulation II. Experiment
Figure 5.14. Induction motor main winding current steady-state waveforms.
Modulation index = 1.375, rotor speed = 1840 rpm.
-3.1
-3.2
Battery Current (A)
-3.3
-3.4
-3.5
-3.6
0.01 0.02 0.03 0.04 0.05
Time (s)
I. Simulation II. Experiment
Figure 5.15. System battery current steady-state waveforms. Modulation index = 1.375,
rotor speed = 1840 rpm.
Figure 5.14 shows the single-phase induction motor main winding current
The system steady-state battery current is shown in Figure 5.15. The simulated
and experimental waveform clearly shows that the battery is being charged (that is the
337
battery is absorbing power). The experimental current waveform in Figure 5.15 also
generator using harmonic balance have been discussed in Chapter 4. Equations 4.52 -
M 2 Rd I dss
MVdo + − rds I dss − jω e (Lds I dss + Lmd I drr
′ )= 0 (5.27)
4(1 − j 2ω e C d Rd )
( ′ =0
Vqss − rqs I qss − jωe Lqs I qss + Lmq I qrr ) (5.28)
− N dq ω r Lqr(
′ I qrr )
′ + Lmq I qss − rdr ′ − jωe (Ldr
′ I drr ′ + Lmd I dss ) = 0
′ I drr (5.30)
4
(
Te1 = P N dq λ ′qrr I drr
′* − N qd λ ′drr I qrr
′* ) (5.31)
1
Cq
( )
I qss + I qssm + jωeVqss = 0 (5.32)
( ′
Vqssm − rqsm I qssm − jωe Lqsm I qssm + Lmqm I qrrm =0 ) (5.33)
N qdm ω r (Ldrm
′ I drrm
′ + Lmdm I dssm ) − rqrm
′ I qrrm
′ − jωe Lqrm
′ I qrrm
′ (
+ Lmqm I qssm = 0) (5.34)
− N dqmω r Lqrm (
′ I qrrm
′ + Lmqm I qssm − rdrm )
′ I drrm
′ − jω e (Ldrm
′ I drrm
′ + Lmdm I dssm ) = 0 (5.35)
Tem =
Pm
4
(
N dqm λ ′qrrm I drrm
′* − N qdm λ ′drrm I qrrm
′* . ) (5.36)
338
The steady-state equations (5.27 - 5.36) are used to solved for the state variables
using MATLAB.
inverter SPIG system feeding a SPIM will be compared with the predicted performance
of the system. The steady state performance was carried at different values of modulation
indexes. The experimental measurement and calculation was carried out with generator
rotor speed of 1840 rpm, a carrier frequency of 2 kHz, and modulating frequency of 60
Hz. The battery voltage and load capacitor are 144 V and 180 µF respectively. The
constant parameters used for the steady-state experiment and calculation are shown in
appendix B.
Figure 5.16: Measured and calculated motor input voltage as a function of motor speed.
Constant generator rotor speed = 1840 rpm.
339
The measured and calculated steady-state input voltage into the single-phase
induction motor as a function of the motor speed is shown in Figure 5.16. Figure 5.16
shows the motor input voltages are higher with the higher modulation index. In addition
Figure 5.17 shows the measured and calculated steady-state torque of the single-
phase induction motor as a function of motor speed. The motor torque decreases with
increase in the motor speed. The figure also indicates that the higher the modulation
Figure 5.18 shows how measured and calculated steady-state input power to the
induction motor varies as a function of the motor speed. It can be observed that the
predicted performance curves show the trend of the actual measured points, but there is
fairly substantial magnitude difference in output power between the predicted and
measured curves.
0.9
0.8 0.8
1.0
0.7 1.5
Torque (Nm)
2.0
0.6 Calc
0.5
0.4
0.3
0.2
0.1
355 360 365 370 375
Motor Speed (rad/s)
Figure 5.17: Measured and calculated motor torque as a function of motor speed.
Constant generator rotor speed = 1840 rpm.
340
350
0.8
1.0
300
1.5
200
150
100
50
355 360 365 370 375
Motor Speed (rad/s)
Figure 5.18: Measured and calculated motor input voltage as a function of motor speed.
Constant generator rotor speed = 1840 rpm.
4.5
0.8
Main Winding Current (A)
1.0
4 1.5
2.0
Calc
3.5
2.5
2
355 360 365 370 375
Motor Speed (rad/s)
Figure 5.19: Measured and calculated generator winding current as a function of motor
speed. Constant generator rotor speed = 1840 rpm.
function of motor speed is shown in Figure 5.19. The main winding current increases
with increase in the modulation index and decreases almost linearly with increases in
341
0.5
-0.5
-1.5
-2
0.8
-2.5 1.0
1.5
-3
2.0
-3.5 Calc
-4
355 360 365 370 375
Motor Speed (rad/s)
Figure 5.20: Measured and calculated battery current as a function of motor speed.
Constant generator rotor speed = 1840 rpm.
motor speed. The predicted curves are similar to the measurement though they are
different in magnitude.
The graph shown Figure 5.20 displays the measured and calculated battery
steady-state current as a function of the motor speed. The battery currents are negative
In this section, the predicted steady state performance of the battery inverter SPIG
system feeding a SPIM load will be discussed. The steady state performance was carried
were carried out with generator rotor speed of 1840 rpm, a carrier frequency of 2 kHz and
modulating frequency of 60 Hz. The battery voltage and load capacitor are 144 V and
342
210 µF, respectively. The constant parameters used for the steady-state experiment and
Figure 5.21 shows the motor voltage as a function of the motor torque at different
modulation index. The motor voltage and torque increases as the modulation index
increases. For a fixed value of modulation index the motor voltage decreases as the motor
Figure 5.22 shows the motor voltage as a function of the motor input power at
different modulation index. The motor voltage and input power increases as the
modulation index increases. For a fixed value of modulation index the motor voltage
90
80
Motor Voltage (V)
70
60
50
M=0.8
M=1.2
40 M=1.6
M=2.0
30
0 0.2 0.4 0.6 0.8 1 1.2
Torque (Nm)
Figure 5.21: Predicted motor voltage as a function of motor torque. Constant generator
rotor speed = 1840 rpm.
343
90
80
60
50
M=0.8
M=1.2
40 M=1.6
M=2.0
30
0 100 200 300 400
Motor Input Power (W)
Figure 5.22: Predicted motor voltage as a function of motor input power. Constant
generator rotor speed = 1840 rpm.
1.2
M=0.8
M=1.2
1 M=1.6
M=2.0
Torque (Nm)
0.8
0.6
0.4
0.2
0
0 100 200 300 400
Motor rotor speed (rad/s)
Figure 5.23: Predicted motor torque as a function of motor rotor speed. Constant
generator rotor speed = 1840 rpm.
344
Figure 5.23 shows the motor torque as a function of the motor rotor speed at
different modulation index. As the modulation index increases the motor torque
increases. In addition as the motor torque increases as the motor rotor speed increases the
maximum torque is obtained at about 330 rad/s of the motor rotor speed then the torque
start falling as the motor rotor speed is increased further. If maximum torque is desired, it
could be obtained by operating the system in the overmodulation region and at a motor
This section gives a description of the influences of the system parameter on the
system performance. A proper selection of the motor rotor speed and the generator rotor
speed can ensure that the system with induction motor operates in optimum condition.
The selection of the motor load rotor speed and the generator rotor speed can be carefully
done to ensure overall system efficiency, maximum torque and maximum power. For
instance, if maximum torque is desired for driving a pump, the appropriate generator and
In the study that follows three load frequencies are selected - 30Hz, 45Hz and
60Hz. The modulation index is fixed at a value of one and the load capacitor, Cq is set
Figure 5.24 shows the contour plot of the motor torque as a function of the
generator rotor speed and the motor rotor speed for different load frequency. As the load
345
frequencies increase the motor torque decrease for constant values of generator rotor
speed and motor rotor speed. In particular at a specific load frequency (45 Hz) maximum
torque can be obtained at high value of motor rotor speed and at about 1.1 p.u. of the
generator rotor speed. The graph will be an excellent tool in selecting generator rotor
speed and motor rotor speed for maximum torque operation such as in pump application.
The contour curves for the motor voltage as a function of the generator rotor
speed and the motor rotor speed for different load frequencies is shown in Figure 5.25. In
general at a constant load frequency the motor voltage is increases at a higher motor rotor
speed and constant generator speed. The graph also indicates that maximum motor
voltage is obtained at generator rotor speed of about 1.12p.u. The graph shows that as the
frequency decreases the motor voltage is relatively constant at constant generator rotor
Figure 5.26 shows the contour plot of generator output power as a funnction of
the generator rotor and the motor rotor speeds for different load frequencies. Considering
a load frequency of 30 Hz for instance the maximum power is obtained at a higher motor
rotor speed and at a generator rotor speed of about 1.18p.u. The graph also indicates that
the choice of operating load frequency generally affects the output power. If the
The family of contour curves of the battery current as a function of the generator
rotor and the motor rotor speeds for different load frequencies is given in Figure 5.27. At
a constant generator rotor speed the graph shows that the battery current is fairly constant
at different values of motor rotor speed. The graph also indicates that the battery current
generally decreases as load frequency increases. The negative battery current indicates
346
the condition when the battery absorbs real power while positive battery current shows
5.6 Transient and Dynamic Performance for the System Feeding a SPIM Load
This section examines the transient and dynamic performance of the battery
transient and dynamic studies give us better understanding of the operation of the system
and also show how the system can perform with either change in load torque or generator
rotor speed. The system start-up process is initially set forth. The response of the system
to changes in load torque is discussed next. The system dynamic performance with the
generator rotor speed changed is discussed and afterward the system performance to a
This section examines the start-up process for the battery inverter generator
system feeding a single-phase induction motor. The simulation was carried out with a
generator rotor speed of 1840 rpm under no load condition. The modulation index was set
a value of 0.80 with a carrier frequency of 2 kHz and modulating frequency of 60 Hz.
347
0.9 0.9
0.8 0.8
0.268 1.26
2.38
0.7 0.7
0.436
wrm/we
wrm/we
0.932
0.6 0.6
1.73
0.5 0.5
0.6
0.4 1.08 0.4
f=30Hz f=45Hz 0.268
0.3 0.3
0.2 0.2
0.8 0.9 1 1.1 1.2 1.3 0.8 0.9 1 1.1 1.2 1.3
wr/we wr/we
0.9
0.856
0.8
0.7 0.635
wrm/we
0.6
0.413
0.5
0.4 0.191
f=60Hz
0.3
0.2
0.8 0.9 1 1.1 1.2 1.3
wr/we
Load Frequency, f = 30Hz Load Frequency, f = 45Hz
Load Frequency, f = 60Hz
Figure 5.24: Contour plot of torque, [Nm] as a variation of generator rotor speed (per
unit) and motor rotor speed (per unit).
348
1 1
57.9
58.2
0.8 0.8
27.9 47.7
0.6 0.6
wrm/we
wrm/we
27.2 37.4
48.1 48.1
0.4 0.4
f=45Hz 37.4
f=30Hz 38
27.2
0.2 0.2
38
0 0
0.8 0.9 1 1.1 1.2 1.3 0.8 0.9 1 1.1 1.2 1.3 1.4
wr/we wr/we
1
61
0.8
49.8 38.6
0.6
wrm/we
38.6
0.4
f=60Hz 27.5
0.2
27.5
0
0.8 0.9 1 1.1 1.2 1.3
wr/we
Load Frequency, f = 30Hz Load Frequency, f = 45Hz
Load Frequency, f = 60Hz
Figure 5.25: Contour plot of motor voltage, [V] as a variation of generator rotor speed
(per unit) and motor rotor speed (per unit).
349
0.9 0.9
151
0.8 0.8
179 30.9
29.6
wrm/we
wrm/we
0.4 0.4
0.8 0.9 1 1.1 1.2 1.3 0.8 0.9 1 1.1 1.2 1.3
wr/we wr/we
0.9
142
0.8
105
wrm/we
0.7
67.9
0.6
f=60Hz 31.1
0.5
0.4
0.8 0.9 1 1.1 1.2 1.3
wr/we
Load Frequency, f = 30Hz Load Frequency, f = 45Hz
Load Frequency, f = 60Hz
Figure 5.26: Contour plot of output power, [W] as a variation of generator rotor speed
(per unit) and motor rotor speed (per unit).
350
1 1
-2.27 4.84
7.26
0.8 0.8
-1.11 -2.27
0.6 0.6
wrm/we
wrm/we
0 0
0.8 0.9 1 1.1 1.2 1.3 0.8 0.9 1 1.1 1.2 1.3 1.4
wr/we wr/we
-2.57
0.8
0.4
f=60Hz 0.649 0.649
0.2
0
0.8 0.9 1 1.1 1.2 1.3
wr/we
Load Frequency, f = 30Hz Load Frequency, f = 45Hz
Load Frequency, f = 60Hz
Figure 5.27: Contour plot of battery current, [A] as a variation of generator rotor speed
(per unit) and motor rotor speed (per unit).
351
0.9 0.9
0.8 0.8
0.268 1.26
2.38
0.7 0.7
0.436
wrm/we
wrm/we
0.932
0.6 0.6
1.73
0.5 0.5
0.6
0.4 1.08 0.4
f=30Hz f=45Hz 0.268
0.3 0.3
0.2 0.2
0.8 0.9 1 1.1 1.2 1.3 0.8 0.9 1 1.1 1.2 1.3
wr/we wr/we
0.9
0.856
0.8
0.7 0.635
wrm/we
0.6
0.413
0.5
0.4 0.191
f=60Hz
0.3
0.2
0.8 0.9 1 1.1 1.2 1.3
wr/we
Load Frequency, f = 30Hz Load Frequency, f = 45Hz
Load Frequency, f = 60Hz
Figure 5.24: Contour plot of torque, [Nm] as a variation of generator rotor speed (per
unit) and motor rotor speed (per unit).
352
1 1
57.9
58.2
0.8 0.8
27.9 47.7
0.6 0.6
wrm/we
wrm/we
27.2 37.4
48.1 48.1
0.4 0.4
f=45Hz 37.4
f=30Hz 38
27.2
0.2 0.2
38
0 0
0.8 0.9 1 1.1 1.2 1.3 0.8 0.9 1 1.1 1.2 1.3 1.4
wr/we wr/we
1
61
0.8
49.8 38.6
0.6
wrm/we
38.6
0.4
f=60Hz 27.5
0.2
27.5
0
0.8 0.9 1 1.1 1.2 1.3
wr/we
Load Frequency, f = 30Hz Load Frequency, f = 45Hz
Load Frequency, f = 60Hz
Figure 5.25: Contour plot of motor voltage, [V] as a variation of generator rotor speed
(per unit) and motor rotor speed (per unit).
353
0.9 0.9
151
0.8 0.8
179 30.9
29.6
wrm/we
wrm/we
0.4 0.4
0.8 0.9 1 1.1 1.2 1.3 0.8 0.9 1 1.1 1.2 1.3
wr/we wr/we
0.9
142
0.8
105
wrm/we
0.7
67.9
0.6
f=60Hz 31.1
0.5
0.4
0.8 0.9 1 1.1 1.2 1.3
wr/we
Load Frequency, f = 30Hz Load Frequency, f = 45Hz
Load Frequency, f = 60Hz
Figure 5.26: Contour plot of output power, [W] as a variation of generator rotor speed
(per unit) and motor rotor speed (per unit).
354
1 1
-2.27 4.84
7.26
0.8 0.8
-1.11 -2.27
0.6 0.6
wrm/we
wrm/we
0 0
0.8 0.9 1 1.1 1.2 1.3 0.8 0.9 1 1.1 1.2 1.3 1.4
wr/we wr/we
-2.57
0.8
0.4
f=60Hz 0.649 0.649
0.2
0
0.8 0.9 1 1.1 1.2 1.3
wr/we
Load Frequency, f = 30Hz Load Frequency, f = 45Hz
Load Frequency, f = 60Hz
Figure 5.27: Contour plot of battery current, [A] as a variation of generator rotor speed
(per unit) and motor rotor speed (per unit).
355
356
357
Figure 5.28 shows the motor input voltage waveform during startup. The motor
input voltage waveform increases quickly until it reaches a steady-state peak value of
about 46 volts. Figure 5.29 shows the generator main winding current. The main winding
Figure 5.28 Motor voltage start-up waveform. Modulation index = 0.8, and generator
rotor speed = 1840 rpm.
Figure 5.30 shows the motor rotor speed transient waveform. The motor rotor
speed ramps gradually and then to a steady state value. The motor speed graph is
observed to change sharply about 4.5 seconds when the auxiliary winding of the motor in
400
350
300
Motor speed (rad/s)
250
200
150
100
50
0
0 1 2 3 4 5 6
Time (s)
Figure 5.30 Motor rotor speed start-up waveform. Modulation index = 0.8, and generator
rotor speed = 1840 rpm.
Figure 5.31 Motor torque start-up waveform. Modulation index = 0.8, and generator rotor
speed = 1840 rpm.
359
The no load motor torque start-up waveform is shown in Figure 5.31. The motor
The generator winding current waveform build-up is shown in Figure 5.32. The
current increases rapidly to a steady state peak value of about 8.0A. Figure 5.33 gives the
transient waveform of the auxiliary winding current. The auxiliary winding current
The transient waveform of the battery current is shown in Figure 5.34 during start
up. The battery current is initially high then decreases to a steady state value. The battery
current is positive indicating that the battery supplies real power to meet the load
5.35. The graph shows how the generator torque grows quickly to a steady state value.
Figure 5.33 Generator auxiliary winding current start-up waveform. Modulation index =
0.8, and generator rotor speed = 1840 rpm.
Figure 5.34 Battery current start-up waveform. Modulation index = 0.8, and generator
rotor speed = 1840 rpm.
361
Figure 5.35 Generator torque start-up waveform. Modulation index = 0.8, and generator
rotor speed = 1840 rpm.
This section examines the dynamics for the battery inverter generator system
feeding a SPIM for changes in the load torque. The simulation was carried out with
generator rotor speed of 1840 rpm and the modulation index was set a value of 2.0 with a
The load torque is initial set at a value of 0 Nm. When the system reaches its
changed to 0.2 Nm. The changes in the load torque are illustrated in Figure 5.36.
Figure 5.37 shows the motor input voltage waveform. The graph shows how the
system responded to a change in load torque. The motor input voltage decreases as the
load torque is increased from 0 Nm to 1.2 Nm. The response of the motor rotor speed to
changes in the load torque is shown in Figure 5.38. The motor rotor speed decreases as a
362
load torque of 1.2 Nm is applied to the motor. When the load torque is reduced to 0.2 Nm
1.2
0.8
0.6
0.4
0.2
0
0.5 0.6 0.7 0.8 0.9 1 1.1
Time (s)
Figure 5.36. Changes in the values of the load torque. Modulation index = 2.0 and rotor
generator speed = 1840 rpm.
Figure 5.37. Motor input voltage waveform response to changes in load torque.
Modulation index = 2.0 and rotor generator speed = 1840 rpm.
363
Figure 5.38. Motor rotor speed waveform response to changes in load torque. Modulation
index = 2.0 and rotor generator speed = 1840 rpm.
Figure 5.39. Motor torque waveform response to changes in load torque. Modulation
index = 2.0 and rotor generator speed = 1840 rpm.
364
Figure 5.40. Generator main winding current waveform response to changes in load
torque. Modulation index = 2.0 and rotor generator speed = 1840 rpm.
Figure 5.41. Generator auxiliary winding current waveform response to changes in load
torque. Modulation index = 2.0 and rotor generator speed = 1840 rpm.
The response of the motor torque waveform to a change in load torque is shown
in Figure 5.39. The motor torque initially increases and then decreases in response to a
change in load torque. Figure 5.40 shows the main winding current waveform response to
365
a change in load torque. The main winding current waveform increases and decreases in
Figure 5.42. Battery current waveform response to changes in load torque. Modulation
index = 2.0 and rotor generator speed = 1840 rpm.
Figure 5.43. Generator torque waveform response to changes in load torque. Modulation
index = 2.0 and rotor generator speed = 1840 rpm.
366
Figure 5.41 shows the auxiliary winding current waveform response to changes in
load torque. The main winding current waveform decreases and increases in response to
changes in load torque. The battery current waveform in response to changes in load
torque is shown in Figure 5.42. The battery current increases and decreases in response to
changes in load torque. The battery current is negative showing that the battery is
absorbing real power from the load. Figure 5.43 shows the generator torque waveform
response to changes in load torque. The generator speed decreases and increases in
This section examines the dynamics for the battery inverter generator system
feeding a SPIM load for changes in the generator rotor speed. The simulation was carried
out with load torque of 0.25Nm. The modulation index was set a value of 2.0 with a
The generator rotor speed is initial is set at a value of 1840 rpm when the system
reaches its steady state operating condition it is changed from 1840 rpm to 1900 rpm then
Figure 5.45 shows motor input voltage waveform response to changes in the
generator rotor speed. It is observed that the motor voltage is still maintained while the
generator rotor speed is less than the 1800 rpm (synchronous speed) as the battery
supplies the needed real power required by the load. Figure 5.46 shows the main winding
367
current waveform. The main winding current decrease as the generator rotor speed
1900
Generator Rotor Speed (rpm)
1850
1800
1750
Figure 5.45 Motor input voltage waveform response to changes in generator rotor speed.
Modulation index = 2.0 and Load Torque = 0.25 Nm.
368
Figure 5.47 shows motor rotor speed waveform in response to changes in the
generator rotor speed. The motor speed decreases slightly in response to changes in
generator rotor speed from 1900 rpm to 1740 rpm. Figure 5.48 shows the motor torque
waveform in response to changes in generator rotor speed. The motor torque increases
Figure 5.49 shows the generator main winding current response to changes in
generator rotor speed. The generator main winding current decreases as the generator
rotor speed changes from 1900 rpm to 1740 rpm. The auxiliary winding current
waveform response to changes in the generator rotor speed is shown in Figure 5.50. The
auxiliary winding current initially increases then decreases in response to changes in the
Figure 5.46 Motor main winding current waveform response to changes in generator
rotor speed. Modulation index = 2.0 and Load Torque = 0.25 Nm.
369
Figure 5.47 Motor rotor speed waveform response to changes in generator rotor speed.
Modulation index = 2.0 and Load Torque = 0.25 Nm.
Figure 5.48 Motor torque waveform response to changes in generator rotor speed.
Modulation index = 2.0 and Load Torque = 0.25 Nm.
370
Figure 5.49 Generator main winding current waveform response to changes in generator
rotor speed. Modulation index = 2.0 and Load Torque = 0.25 Nm.
Figure 5.51 shows the battery current response to changes in generator rotor
speed. The battery current increases as the generator rotor speed changes from 1900 rpm
to 1740 rpm. The graph also shows that the system is able to sustain its operation even
when the generator rotor speed (1740 rpm) is less than the synchronous speed (1800 rpm)
as the battery provides real power requirement for the motor load.
The generator torque waveform response to changes in the generator rotor speed
is shown in Figure 5.52. The generator torque increases in response to changes in the
generator rotor speed from 1900 rpm to 1740 rpm. It can be observed that the generator
Figure 5.51. Battery current waveform response to changes in generator rotor speed.
Modulation index = 2.0 and Load Torque = 0.25 Nm.
372
Figure 5.52 Generator torque waveform response to changes in generator rotor speed.
Modulation index = 2.0 and Load Torque = 0.25 Nm.
This section examines the dynamics of the battery inverter generator system
feeding a SPIM when heavily loaded. The simulation was carried out with generator rotor
speed of 1840 rpm. The modulation index was set a value of 1.2 with a carrier frequency
The load torque is initial set at a value of 0 Nm when the system reaches its
steady state operating condition it is changed from 0 Nm to 4.0 Nm. The changes in load
Figure 5.54 shows the motor input voltage waveform. The graph shows how the
system response to a change in load torque. The motor input voltage collapses to a lower
peak value as it is loaded with a torque of 4 Nm. The response of the motor main winding
373
current to a change in the load torque is shown in Figure 5.55. The motor main winding
2.5
1.5
0.5
0
0.4 0.45 0.5 0.55 0.6 0.65
Time (s)
Figure 5.53. Changes in the values of the load torque. Modulation index = 1.2
and rotor generator speed = 1840 rpm.
Figure 5.54. Motor input voltage waveform. Modulation index = 1.2 and
rotor generator speed = 1840 rpm.
374
Figure 5.55. Motor main winding current waveform. Modulation index = 1.2 and
rotor generator speed = 1840 rpm.
Figure 5.56. Motor speed waveform. Modulation index = 1.2 and rotor generator
speed = 1840 rpm.
375
Figure 5.57. Motor torque waveform. Modulation index = 1.2 and rotor generator
speed = 1840 rpm.
Figure 5.58. Generator main winding current waveform. Modulation index = 1.2 and
rotor generator speed = 1840 rpm.
376
Figure 5.59. Generator auxiliary winding current waveform. Modulation index = 1.2 and
rotor generator speed = 1840 rpm.
The response of the motor rotor speed to a change in the load torque is shown in
Figure 5.56. The graph shows that the motor rotor speed falling rapidly as it is loaded
with a load of 4 Nm. Figure 5.57 shows the motor torque waveform. As the system is
loaded with 4 Nm the motor stalls as show in the waveform (torque reduces).
The generator main winding current is shown in Figure 5.58. The graphs shows
that the current grows quickly as it is loaded. Figure 5.59 shows the waveform of the
generator auxiliary winding current. The auxiliary winding current grows rapidly as the
system is loaded.
Figure 5.60. The graph indicates that initial the battery absorbs real power (negative
current), when it is loaded afterwards the current increases to meet the power requirement
of the motor. Figure 5.61 shows the system generator torque waveform response to a
change in load torque. The waveform grows quickly as the system is loaded.
377
Figure 5.60. Battery current waveform. Modulation index = 1.2 and rotor generator
speed = 1840 rpm.
Figure 5.61. Generator torque waveform. Modulation index = 1.2 and rotor generator
speed = 1840 rpm.
378
CHAPTER 6
6.1 Introduction
generator system feeding an impedance load will be presented. The block diagram of the
single-phase generator system without battery and impedance load is shown in Figure
6.1. This generator system is similar to the system in Figure 4.1. The only difference is
that the battery is not connected. The operation of the capacitor inverter SPIG system is
The first part of this chapter deals with the development of the steady state
mathematical model for the generator system feeding an impedance load using harmonic
balance technique. Next the condition for self-excitation for the generator system without
battery is discussed. The minimum requirements for system parameters for fixed
Vd
Single
+ PWM SPIG phase
Cd Inverter
Impedance
-
Load
173
379
The PWM inverter, the single-phase induction generator, and the impedance load
makes up the generator system. The schematic diagram of the generator system is shown
in Figure 6.2.
The capacitor inverter SPIG system operation is similar to the generator system
description in section 4.2. The main difference is that capacitor inverter SPIG system
under consideration is without the battery. The full charged capacitor is connected into
the input of the PWM inverter. The inverter is switched using the bipolar voltage
switching scheme. The output of the inverter is fed to the auxiliary winding of the single-
phase generator. The output from the main winding is fed to the impedance load. The
system has the same features as the system discussed in chapter four. When the SPIG is
supplied more than enough power required by the motor load the excess power is used to
charge the capacitor and when the power provided by the generator is less than the
Load
R o Lo
TA+ TB+
Cq
Aux. Winding
Vd
Cd
Main Winding
TA- TB-
requirement by the impedance load the capacitor discharges to provide the balance. A
index.
The equivalent circuit of the system that is the capacitor, PWM inverter, single-
The dynamic equation for the capacitor-PWM inverter generator system is given
as
pVd =
1
(− I ds S a ) (6.1)
Cd
pλ ds = V ds − rds I ds (6.3)
pλ qs = V qs − rqs I qs (6.4)
Figure 6.3. The q-d equivalent circuit of the capacitor-PWM inverter generator system
with impedance load. (a) d-axis, (b) q-axis circuit.
381
1 V
pVqs = − I qs + qs (6.7)
Cq Zo
where the capacitor voltage and current are denoted Vd and Id. The other state variable
Equations 4.12 - 4.15 representing the q-d flux linkages of the generator are given
as
′
λ qs = L qs I qs + Lmq I qr (6.8)
′
λ ds = Lds I ds + Lmd I dr (6.9)
λ ′qr = L qr
′ I qr
′ + Lmq I qs (6.10)
λ ′dr = Ldr
′ I dr
′ + Lmd I ds . (6.11)
The dynamics of the generator turbine is given in Equations (4.16 and 4.17) and
are given as
2
(
Te = P N dq λ ′qr I dr
′ − N qd λ ′dr I qr
′ ) (6.13)
The capacitor inverter generator system in Figure 6.2 does not have any battery
source; for excitation to occur the system need to be carefully studied. If a close
examination of the generator system equation is carried out the necessary parameter for
technique was used in determine the condition for self-excitation. The steady state
equation can be obtained from the dynamic Equation 6.1 to 6.13. Sa, Ids and Vd are
defined as
Sa = M q cos( θ ) + M dq sin( θ )
pVd =
1
Cd
( )
M q cos( θ ) + M dq sin( θ ) * (I ds1 cos( θ ) + I ds 2 sin( θ )) ; (6.14)
Since Equations 6.13 and 6.14 are equal, terms must also be equal. Comparing terms in
pVdo = −
1
2Cd
(
M q I ds1 + M d I ds 2 ) (6.16)
pVd 1 + 2ωeVd 2 = −
1
2Cd
(
M q I ds1 − M d I ds 2 ) (6.17)
383
pVd 2 − 2ωeVd 1 = −
1
2Cd
(
M q I ds 2 + M d I ds1 . ) (6.18)
1 1 1 1
Vds = M qVdo + M qVd 1 + M dVd 2 cos( θ ) + M dVdo + M qVd 2 − M dVd 1 sin( θ ) ;
2 2 2 2
(6.19)
Vds = (rds I ds1 + ωeλ ds 2 + pλ ds1 ) cos (θ) + (rds I ds 2 − ωeλ ds1 + pλ ds 2 ) sin(θ) . (6.20)
Since Equations 6.19 and 6.20 are equal, the terms are also equal. Hence
1 1
rds I ds1 + ωeλ ds 2 + pλ ds1 = M qVdo + M qVd 1 + M dVd 2 (6.21)
2 2
1 1
rds I ds 2 − ωeλ ds1 + pλ ds 2 = M dVdo + M qVd 2 − M dVd 1 . (6.22)
2 2
With harmonic balance technique applied to Equations (6.4 - 6.7) the following equation
results
1 V
p Vqs1 + ωeVqs 2 = − I qs1 + qs1 (6.29)
Cq Zo
384
1 V
p Vqs2 − ωeVqs1 = − I qs 2 + qs 2 . (6.30)
Cq Zo
At steady state, peak of the state variables are constant making derivatives in Equations
The system equation in terms of flux is then obtained by substituting for current
using Equations 6.8 - 6.11. The system equation in matrix form is given in Equation 6.31.
385
iy
0
i 0
0
0
0
0
2 we
0
0
Cdd Mq U2
Cdd Md U2
Cdd Md U2
- 2 we Cdd Mq U2 - Cdd Md U2
Cdd Mq U2
0
0
0
0
0
0
0
0
0
0
0
0
- Cdd Mq U1 - Cdd Md U1
- Cdd Mq U1 Cdd Md U1
- Cdd Md U1 - Cdd Mq U1
0
0
0
0
0
0
yi y
Vdo
0 Vd1
Mq Md
0 Mq U4 - we 0 0 0 0 - U3 0 0 0 Vd2
2 2
0 Mq l ds1
Md
0 Md - we U4 0 0 0 0 0 - U3 0 0 l ds2
2 2
0 0 0 0 0 0 - U6 - we U5 0 0 0 1 0 l qs1
0 = 0 0 0 0 0 we - U6 0 - U5 0 0 0 1 l qs2
0 0 0 0 0 0 U7 0 - U8 - we Nqd wr 0 0 0 l qr1
0 0 0 0 0 0 0 U7 we - U8 0 Nqd wr 0 0 l qr2
0 wr l dr1
0 0 0 - U9 0 0 0 - Nqd 0 T1 - we 0 0
k{k k {
0 l dr2
wr
0 0 0 0 0 - U9 0 0 0 - we T1 0 0 Vqs1
{
Nqd
0 Vqs2
0 0 0 0 0 - Cqq T3 0 Cqq T2 0 0 0 - Cqq Zoo - we
0 0 0 0 0 0 - Cqq T3 0 Cqq T2 0 0 we - Cqq Zoo
(6.31)
Lmd 1 L r L r L rqs Lmq rqs Lqr rqr Lmq rqr Lqs
where U1 = , Cdd = , U 2 = dr , U 3 = ds md , U 4 = ds dr , U 5 = , U6 = , U7 = , U8 = ,
2∆ 2 C 2∆ 2 ∆2 ∆2 ∆1 ∆1 ∆1 ∆1
U9 =
rdr Lmd
∆2
r L
, T1 = dr ds , T2 =
∆2
Lmq
∆1
, T3 =
Lqr
∆1
, Cqq =
1
Cq
, Z oo =
1
Zo
( ) (
, ∆1 = Lqs Lqr − L2mq and ∆ 2 = L2md − Lds Ldr . )
Equation 6.31 is of the form [0] = [A][X ] where [X] is the state variables. The solution to this equation is relatively straight
forward, it is either that [A] = 0 or [X] = 0, since the state cannot be zero hence the solution is therefore [A] = 0. With this, the [A]
matrice can be solved for the condition for self-excitation. The condition for self-excitation is found by evaluating the determinant of
6.4. Results
The evaluating [A] matrix determinant clearly shows that for self-excitation to occur it is independent of the input capacitor,
Cd. The effect of change in parameter is a secondary effect [16] in self-excitation of the generator system hence constant parameters
The results that follows are obtained by assuming that Md is zero that is modulation index is real in other words the angle is
zero. The determinant of A matrices when Md=0 shows that the generator system is independent of the modulation index, Mq. Figure
180
6.4 shows the required value of load resistance to choose at a corresponding rotor speed. The load frequency and capacitor, Cq are kept
constant.
80
70
60
Load Resistor (Ohms)
50
40
30
20
10
0
380 390 400 410 420
Rotor Speed (rad/s)
Figure 6.4. Self-excitation for the system without battery. Calculated load resistance vs
generator rotor speed.
In order to achieve a self-excitation at a load capacitor, Cq of 180 µF and load frequency of 60Hz, and generator rotor speed ωr
of 385 rad/s will require a minimum load impedance of 14.72 Ohms as show in Figure 6.4.
181
Figure 6.5 shows the contour plot at various value of load resistance. From Figure 6.5 given specific load impedance, the
corresponding load capacitor and generator rotor speed for self-excitation of the system could be obtained. Under this condition load
300
50
250 40
Load Capacitor (uF)
200 30
150
20
10
100
The minimum load impedance to obtain self-excitation with variation of Md and Mq was also computed. Figure 6.6 shows the
required value of load resistance to choose at a corresponding value of Md and Mq. The load frequency, generator rotor speed, and
capacitor, Cq are kept constant. The constant parameter value are the load frequency, f = 60Hz, generator rotor speed, ωr = 1840 rpm,
Figure 6.6 shows the contour plot at various value of load resistance. From Figure 6.6 given specific load impedance, the
corresponding modulation Md and Mq for self-excitation of the system could be obtained with other system parameter kept constant.
183
1
29.2 23.1
0.8 26.2 17
20.1
13.9
0.6
Md
0.4
0.2
0
0 0.1 0.2 0.3 0.4 0.5
Mq
Figure 6.6. Self-excitation for the system without battery. Contour plot of load resistance, [Ω] as a variation of Md and Mq .
184
iy
0
i 0
0
0
0
0
2 we
0
0
Cdd Mq U2
Cdd Md U2
Cdd Md U2
- 2 we Cdd Mq U2 - Cdd Md U2
Cdd Mq U2
0
0
0
0
0
0
0
0
0
0
0
0
- Cdd Mq U1 - Cdd Md U1
- Cdd Mq U1 Cdd Md U1
- Cdd Md U1 - Cdd Mq U1
0
0
0
0
0
0
yi y
Vdo
0 Vd1
Mq Md
0 Mq U4 - we 0 0 0 0 - U3 0 0 0 Vd2
2 2
0 Mq l ds1
Md
0 Md - we U4 0 0 0 0 0 - U3 0 0 l ds2
2 2
0 0 0 0 0 0 - U6 - we U5 0 0 0 1 0 l qs1
0 = 0 0 0 0 0 we - U6 0 - U5 0 0 0 1 l qs2
0 0 0 0 0 0 U7 0 - U8 - we Nqd wr 0 0 0 l qr1
0 0 0 0 0 0 0 U7 we - U8 0 Nqd wr 0 0 l qr2
0 wr l dr1
0 0 0 - U9 0 0 0 - Nqd 0 T1 - we 0 0
k{k k {
0 l dr2
wr
0 0 0 0 0 - U9 0 0 0 - we T1 0 0 Vqs1
{
Nqd
0 Vqs2
0 0 0 0 0 - Cqq T3 0 Cqq T2 0 0 0 - Cqq Zoo - we
0 0 0 0 0 0 - Cqq T3 0 Cqq T2 0 0 we - Cqq Zoo
(6.31)
CHAPTER 7
7.1 Conclusions
source of excitation was tested and analyzed. The battery included in the system was
found to be an excellent configuration for a source or sink of real power depending on the
given load of the single-phase induction generator. The system modeling was done
recorded for different impedance and motor loads for a one horse power single-phase
induction generator.
The steady-state model for the system was developed for impedance and motor
loads using the harmonic balance technique. Experimental results for different impedance
and motor loads compare favorably with the steady-state model calculation.
inverter was done using device model. Matlab/Simulink was found to be an excellent tool
in modeling model the PWM inverter and the single-phase induction generator.
Simulation results of different impedance and motor loads compare favorably with
185
7.2 Suggestion for Further Work
Proper design of the induction machine as a generator will greatly improve its
performance; for instance designing the induction machine to have appropriate turns for
induction generator scheme with PWM inverter. This configuration will greatly improve
overall system performance since the solar cell will provide additional source of real
power.
ensure controlled load voltage and frequency can be further provided for the system. A
digital signal processing (DSP) implementation of the control scheme will ensure system
191
APPENDIX A
The constant parameters of the 1 hp single-phase generator and a banks of batteries are:
APPENDIX B
bk = (c−k − ck ) .
1
(C.10)
j
In this case k=1
1 m 1 m
a1 = jVd + a − jVd + a = 0 (C.11)
π 4 π 4
1 1 m 1 m 1 m
b1 = − jVd + a − jVd + a = −2Vd + a (C.12)
j π 4 π 4 π 4
1 m 2 m
M = 2 + a = + a (C.13)
π 4 π 2
The real Fourier expansion can also be applied to other conditions of k and n.
REFERENCES
187
1. F. P. de Mello, J. W. Feltes, L. N. Hannett and J. C. White, “Application of induction
generators in power systems,” IEEE Transactions on Power Apparatus and Systems,
Vol. PAS-101, no. 9, pp. 3385-3393, September 1982.
14. M. B. Brennen and A. Abbondanti, "Static Exciters for Induction Generator," IEEE
Trans. on Industry Applications, vol. IA-13, no. 5, pp. 422-428, September/October
1977.
15. J. Arrillaga and D. B. Watson, "Static power conversion from self-excited induction
generators," Proc. IEE, vol. 125, no. 8, pp. 743-746, August 1978.
17. E. Muljadi and T. A. Lipo, “Series compensated PWM inverter with battery supply
applied to an isolated induction generator,” IEEE Trans. on Industry Applications,
vol. 30, no. 4, pp. 1073-1082, July/August 1994.
19. Leon O. Chua, Charles A. Desoer and Ernest S. Kuh, “Linear and Nonlinear
Circuits", McGraw-Hill Book Company, 1987.
20. Mathcad User's Guide, “MathSoft, Inc.", 101 Main Street, Cambridge, MA 02142-
1521, 1995.
21. Mathematical for Student, “Wolfram Research, Inc.", 100 Trade Center Drive
Champaign, IL 61820, USA, 1994.
25. B. K. Bose, Power Electronics and Variable Frequency Drives, Technology and
Applications, IEEE Press, New York, 1997.
26. Semikron International, “SEMIKRON Semidriver Medium power six IGBT and
MOSFET driver SKHI60, SKHI60H4 Preliminary data,” Semikron International,
Sigmundstr. 200, Postfach 82 02 51, D-90253 Nurnberg Germany.
27. International Rectifier, “Designer's Manual Insulated Gate Bipolar Transistor,”
International Rectifier, 233 Kansas St., El Segundo, California 90245, 1996.
28. O. Ojo, Class notes on ECE662 (Advance Electric Machinery), Tennessee Tech. Uni.,
Cookeville, TN, 1996.
31. C. F. Lu, C. C. Liu and C. J. Wu, “Dynamic modelling of battery energy storage
system and application to power system stability,” IEE Proceedings—Generation,
Transmission and Distribution, vol. 142, no. 4, July 1995, pp. 429-435.
32. E. Muljadi, “Series compensated PWM inverter with battery supply applied to an
isolated operation of induction generator,” Ph.D. thesis, University of Wisconsin,
Madison, 1987.
34. Erwin Kreyszig, Advanced Engineering Mathematics, John Wiley and Sons, Inc,
New York, 1979.